WO2013088457A1

WO2013088457A1 - Genetic variants useful for risk assessment of thyroid cancer

Info

Publication number: WO2013088457A1
Application number: PCT/IS2012/050016
Authority: WO
Inventors: Patrick Sulem; Julius Gudmundsson
Original assignee: Decode Genetics Ehf; Illumina Inc
Priority date: 2011-12-13
Filing date: 2012-12-13
Publication date: 2013-06-20

Abstract

Genetic variants conferring risk of thyroid cancer have been discovered. The invention describes the use of such variants in methods of disease management of thyroid cancer. This includes methods of determining susceptibility to thyroid cancer, as well as prognostic methods for thyroid cancer. Also described are kits useful in the methods of the invention.

Description

GENETIC VARIANTS USEFUL FOR RISK ASSESSMENT OF

THYROID CANCER

INTRODUCTION

Thyroid carcinoma is the most common classical endocrine malignancy, and its incidence has been rising rapidly in the US as well as other industrialized countries over the past few decades. Thyroid cancers are classified histologically into four groups: papillary, follicular, medullary, and undifferentiated or anaplastic thyroid carcinomas (DeLellis, R. A., J Surg Oncol, 94, 662 (2006)). In 2008, it is expected that over 37,000 new cases will be diagnosed in the US, about 75% of them being females (the ratio of males to females is 1 :3.2) (Jemal, A., et al., Cancer statistics, 2008. CA Cancer J Clin, 58: 71-96, (2008)) . If diagnosed at an early stage, thyroid cancer is a well manageable disease with a 5-year survival rate of 97% among all patients, yet about 1,600 individuals were expected to die from this disease in 2008 in the US (Jemal, A., et al., Cancer statistics, 2008. CA Cancer J Clin, 58: 71-96, (2008)) . Survival rate is poorer (~40%) among individuals that are diagnosed with a more advanced disease; i.e. individuals with large, invasive tumors and/or distant metastases have a 5-year survival rate of ^;»40% (Sherman, S. I., et al., 3rd, Cancer, 83, 1012 (1998), Kondo, T., Ezzat, S., and Asa, S. L, Nat Rev Cancer, 6, 292 (2006)). For radioiodine-resistant metastatic disease there is no effective treatment and the 10- year survival rate among these patients is less than 15% (Durante, C, et al., J Clin Endocrinol Metab, 91, 2892 (2006)).

Although relatively rare (1% of all malignancies in the US), the incidence of thyroid cancer more than doubled between 1984 and 2004 in the US (SEER web report; Ries L, Melbert D, Krapcho M et al (2007) SEER cancer statistics review, 1975-2004. National Cancer Institute, Bethesda, MD, http://seer.cancer.gov/csr/1975_2004/, based on November 2006 SEER data submission). Between 1995 and 2004, thyroid cancer was the third fastest growing cancer diagnosis, behind only peritoneum, omentum, and mesentery cancers and "other" digestive cancers [SEER web report]. Similarly dramatic increases in thyroid cancer incidence have also been observed in Canada, Australia, Israel, and several European countries (Liu, S., et al., Br J Cancer, 85, 1335 (2001), Burgess, J. R., Thyroid, 12, 141 (2002), Lubina, A., et al., Thyroid, 16, 1033 (2006), Colonna, M., et al., Eur J Cancer, 38, 1762 (2002), Leenhardt, L, et al., Thyroid, 14, 1056 (2004), Reynolds, R. M., et al., Clin Endocrinol (Oxf), 62, 156 (2005), Smailyte, G., et al., BMC Cancer, 6, 284 (2006)).

Thus, there is a need for better understanding of the molecular causes of thyroid cancer progression, to develop new diagnostic tools and better treatment options. The present invention provides thyroid cancer susceptibility variants and their use in various diagnostic applications.

SUMMARY OF THE INVENTION

The present invention relates to methods of risk management of thyroid cancer, based on the discovery that certain genetic variants are correlated with risk of thyroid cancer. Thus, the invention includes methods of determining an increased susceptibility or increased risk of thyroid cancer, as well as methods of determining a decreased susceptibility of thyroid cancer, through evaluation of certain markers that have been found to be correlated with susceptibility of thyroid cancer in humans. The invention also relates to methods of assessing prognosis of individuals diagnosed with thyroid cancer.

In one aspect, the invention relates to method of determining a susceptibility to thyroid cancer, the method comprising analyzing data about at least one allele of human RALGDS gene (SEQ ID NO: 57) in a human subject, wherein different alleles of the human RALGDS gene are associated with different susceptibilities to thyroid cancer in humans, and determining a susceptibility to thyroid cancer for the human subject from the data.

In another aspect, the invention relates to a method of determining a susceptibility to thyroid cancer in a human subject, the method comprising determining the presence or absence of at least one allele of at least one polymorphic marker selected from the group consisting of the markers rsl l3532379, and, and markers in linkage disequilibrium therewith, in a nucleic acid sample obtained from the subject, wherein the presence of the at least one allele is indicative of a susceptibility to thyroid cancer.

The invention also relates to a method of determining whether a human subject is at increased risk of developing thyroid cancer, the method comprising steps of (1) obtaining a biological sample containing nucleic acid from the subject; (2) determining, in the biological sample, nucleic acid sequence about the human RALGDS gene; and (3) comparing the sequence information to the wild-type sequence of RALGDS (SEQ ID NO: 57); wherein an identification of a mutation in RALGDS in the subject is indicative that the individual is at increased risk of developing thyroid cancer.

In another aspect the invention further relates to a method of determining whether a human subject is at increased risk of developing thyroid cancer, the method comprising analyzing amino acid sequence data about a RALGDS polypeptide from the subject, wherein a determination of the presence of a RALGDS polypeptide with altered sequence compa red with a wild-type RALGDS polypeptide with sequence as set forth in SEQ ID NO: 58 is indicative that the subject is at increased risk of developing thyroid cancer.

The invention further relates to a method for determining a susceptibility to thyroid cancer in a human subject, comprising determining the presence or absence of at least one allele of at least one polymorphic marker in a nucleic acid sample obtained from the subject, wherein the at least one polymorphic marker is selected from the group consisting of (a) polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58; and (b) rsl2407041, rs6689698, rsl2136101, rsl 1586476, rs62174266, rs76839330, rsl354833, rs75825480, rs6828277, rsl7540362, rsl060412, rs55635625, rs497341, rs73205431, rs8021657, rsl46663071, rs28524987 and rs62223910; and determining a susceptibility to thyroid cancer from the presence or absence of the at least one allele, wherein the presence of the at least one allele is indicative of a susceptibility to thyroid cancer. The invention also provides an assay for determining a susceptibility to thyroid cancer in a human subject, the assay comprising steps of: (i) obtaining a nucleic acid sample from the human subject; (ii) assaying the nucleic acid sample to determine the presence or absence of at least one allele of at least one polymorphic marker associated with increased susceptibility to thyroid cancer in humans, and (iii) determining a susceptibility to thyroid cancer for the human subject from the presence or absence of the at least one allele, wherein the at least one polymorphic marker is selected from the group consisting of (a) polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58; and (b) rsl2407041, rs6689698, rsl2136101, rsl l586476, rs62174266, rs76839330, rsl354833, rs75825480, rs6828277, rsl7540362, rsl060412, rs55635625, rs497341, rs73205431, rs8021657, rsl46663071, rs28524987 and rs62223910; and wherein a susceptibility to thyroid cancer is determined from the presence or absence of the at least one allele.

The invention also provides kits. In one such aspect, the invention relates to a kit for assessing susceptibility to Thyroid Cancer in human individuals, the kit comprising reagents for selectively detecting at least one at-risk variant for Thyroid Cancer in the individual, wherein the at least one at-risk variant is a polymorphic marker selected from the group consisting of (a)

polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58; and (b) rsl2407041, rs6689698, rsl2136101, rsl l586476, rs62174266, rs76839330, rsl354833, rs75825480, rs6828277, rsl7540362, rsl060412, rs55635625, rs497341, rs73205431, rs8021657, rsl46663071, rs28524987 and rs62223910, and a collection of data comprising correlation data between the at least one at-risk variant and susceptibility to Thyroid Cancer.

The invention also provides computer-implemented applications. In one such application, the invention relates to system for identifying susceptibility to thyroid cancer in a human subject, the system comprising (i) at least one processor; (ii) at least one computer-readable medium; (iii) a susceptibility database operatively coupled to a computer-readable medium of the system and containing population information correlating the presence or absence of one or more alleles of the human RALGDS gene and susceptibility to thyroid cancer in a population of humans; (iv) a measurement tool that receives an input about the human subject and generates information from the input about the presence or absence of at least one mutant RALGDS allele indicative of a RALGDS defect in the human subject; and (v) an analysis tool that (a) is operatively coupled to the susceptibility database and the measurement tool, (b) is stored on a computer-readable medium of the system, (c) is adapted to be executed on a processor of the system, to compare the information about the human subject with the population information in the susceptibility database and generate a conclusion with respect to susceptibility to thyroid cancer for the human subject.

It should be understood that all combinations of features described herein are contemplated, even if the combination of feature is not specifically found in the same sentence or paragraph herein. This includes in particular the use of all markers disclosed herein, alone or in combination, for analysis individually or in haplotypes, in all aspects of the invention as described herein.

BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of preferred embodiments of the invention.

FIG 1 provides a diagram illustrating a system comprising computer implemented methods utilizing risk variants as described herein.

FIG 2 shows an exemplary system for determining risk of thyroid cancer as described further herein.

FIG 3 shows a system for selecting a treatment protocol for a subject diagnosed with thyroid cancer.

FIG 4 depicts a multi-species alignment of RALGDS amino acid sequences from H. sapiens (SEQ ID NO: 58), C.lupus (SEQ ID NO:64), M.musculus (SEQ ID NO:62), R.norvegicus (SEQ ID NO:63), B. taurus (SEQ ID NO: 59), D. melanogaster (SEQ ID NO:60) and G.gallus (SEQ ID NOs:61). Symbols below the sequence alignment highlight residues that are fully (*) or partially ( : or .) conserved between the species.

DETAILED DESCRIPTION

Definitions

Unless otherwise indicated, nucleic acid sequences are written left to right in a 5' to 3' orientation. Numeric ranges recited within the specification are inclusive of the numbers defining the range and include each integer or any non-integer fraction within the defined range. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by the ordinary person skilled in the art to which the invention pertains. The following terms shall, in the present context, have the meaning as indicated :

A "polymorphic marker", sometime referred to as a "marker", as described herein, refers to a genomic polymorphic site. Each polymorphic marker has at least two sequence variations characteristic of particular alleles at the polymorphic site. Thus, genetic association to a polymorphic marker implies that there is association to at least one specific allele of that particular polymorphic marker. The marker can comprise any allele of any variant type found in the genome, including SNPs, mini- or microsatellites, translocations and copy number variations (insertions, deletions, duplications). Polymorphic markers can be of any measurable frequency in the population. For mapping of disease genes, polymorphic markers with population frequency higher than 5-10% are in general most useful. However, polymorphic markers may also have lower population frequencies, such as 1-5% frequency, or even lower frequency, in particular copy number variations (CNVs). The term shall, in the present context, be taken to include polymorphic markers with any population frequency. An "allele" refers to the nucleotide sequence of a given locus (position) on a chromosome. A polymorphic marker allele thus refers to the composition (i.e., sequence) of the marker on a chromosome. Genomic DNA from an individual contains two alleles (e.g., allele-specific sequences) for any given polymorphic marker, representative of each copy of the marker on each chromosome. Sequence codes for nucleotides used herein are: A = 1, C = 2, G = 3, T = 4. For microsatellite alleles, the CEPH sample (Centre d'Etudes du Polymorphisme Humain, genomics repository, CEPH sample 1347-02) is used as a reference, the shorter allele of each microsatellite in this sample is set as 0 and all other alleles in other samples are numbered in relation to this reference. Thus, e.g., allele 1 is 1 bp longer than the shorter allele in the CEPH sample, allele 2 is 2 bp longer than the shorter allele in the CEPH sample, allele 3 is 3 bp longer than the lower allele in the CEPH sample, etc., and allele -1 is 1 bp shorter than the shorter allele in the CEPH sample, allele -2 is 2 bp shorter than the shorter allele in the CEPH sample, etc.

Sequence conucleotide ambiguity as described herein, including sequence listing, is as proposed by IUPAC-IUB. These codes are compatible with the codes used by the EMBL, GenBank, and PIR databases.

A nucleotide position at which more than one sequence is possible in a population (either a natural population or a synthetic population, e.g., a library of synthetic molecules) is referred to herein as a "polymorphic site".

A "Single Nucleotide Polymorphism" or "SIMP" is a DNA sequence variation occurring when a single nucleotide at a specific location in the genome differs between members of a species or between paired chromosomes in an individual. Most SNP polymorphisms have two alleles. Each individual is in this instance either homozygous for one allele of the polymorphism (i.e. both chromosomal copies of the individual have the same nucleotide at the SNP location), or the individual is heterozygous (i.e. the two sister chromosomes of the individual contain different nucleotides). The SNP nomenclature as reported herein refers to the official Reference SNP (rs) ID identification tag as assigned to each unique SNP by the National Center for Biotechnological Information (NCBI).

A "variant", as described herein, refers to a segment of DNA that differs from the reference DNA. A "marker" or a "polymorphic marker", as defined herein, is a variant. Alleles that differ from the reference are referred to as "variant" alleles.

A "microsatellite" is a polymorphic marker that has multiple small repeats of bases that are 2-8 nucleotides in length (such as CA repeats) at a particular site, in which the number of repeat lengths varies in the general population. An "indel" is a common form of polymorphism comprising a small insertion or deletion that is typically only a few nucleotides long.

A "haplotype," as described herein, refers to a segment of genomic DNA that is characterized by a specific combination of alleles arranged along the segment. For diploid organisms such as humans, a haplotype comprises one member of the pair of alleles for each polymorphic marker or locus along the segment. In a certain embodiment, the haplotype can comprise two or more alleles, three or more alleles, four or more alleles, or five or more alleles. Haplotypes are described herein in the context of the marker name and the allele of the marker in that haplotype, e.g., "4 rsl 13532379" refers to the 4 allele of marker rsl 13532379 being in the haplotype, and is equivalent to "rsl 13532379 allele 4". Furthermore, allelic codes in haplotypes are as for individual markers, i.e. 1 = A, 2 = C, 3 = G and 4 = T.

The term "susceptibility", as described herein, refers to the proneness of an individual towards the development of a certain state (e.g. , a certain trait, phenotype or disease), or towards being less able to resist a particular state than the average individual. The term encompasses both increased susceptibility and decreased susceptibility. Thus, particular alleles at polymorphic markers and/or haplotypes of the invention as described herein may be characteristic of increased susceptibility (i.e., increased risk) of thyroid cancer, as characterized by a relative risk (RR) or odds ratio (OR) of greater than one for the particular allele or haplotype. Alternatively, the markers and/or haplotypes of the invention are characteristic of decreased susceptibility (i.e., decreased risk) of thyroid cancer, as characterized by a relative risk of less than one.

The term "and/or" shall in the present context be understood to indicate that either or both of the items connected by it are involved. In other words, the term herein shall be taken to mean "one or the other or both".

The term "look-up table", as described herein, is a table that correlates one form of data to another form, or one or more forms of data to a predicted outcome to which the data is relevant, such as phenotype or trait. For example, a look-up table can comprise a correlation between allelic data for at least one polymorphic marker and a particular trait or phenotype, such as a particular disease diagnosis, that an individual who comprises the particular allelic data is likely to display, or is more likely to display than individuals who do not comprise the particular allelic data. Look-up tables can be multidimensional, i.e. they can contain information about multiple alleles for single markers simultaneously, or they can contain information about multiple markers, and they may also comprise other factors, such as particulars about diseases diagnoses, racial information, biomarkers, biochemical measurements, therapeutic methods or drugs, etc.

A "computer-readable medium", is an information storage medium that can be accessed by a computer using a commercially available or custom-made interface. Exemplary computer- readable media include memory (e.g., RAM, ROM, flash memory, etc.), optical storage media (e.g., CD-ROM), magnetic storage media (e.g., computer hard drives, floppy disks, etc.), punch cards, or other commercially available media. Information may be transferred between a system of interest and a medium, between computers, or between computers and the computer- readable medium for storage or access of stored information. Such transmission can be electrical, or by other available methods, such as IR links, wireless connections, etc.

A "nucleic acid sample" as described herein, refers to a sample obtained from an individual that contains nucleic acid (DNA or RNA). In certain embodiments, i.e. the detection of specific polymorphic markers and/or haplotypes, the nucleic acid sample comprises genomic DNA. Such a nucleic acid sample can be obtained from any source that contains genomic DNA, including a blood sample, sample of amniotic fluid, sample of cerebrospinal fluid, or tissue sample from skin, muscle, buccal or conjunctival mucosa, placenta, gastrointestinal tract or other organs.

The term "thyroid cancer-associated nucleic acid", as described herein, refers to a nucleic acid that has been found to be associated to thyroid cancer. This includes, but is not limited to, the markers and haplotypes described herein and markers and haplotypes in strong linkage disequilibrium (LD) therewith. In one embodiment, a thyroid cancer-associated nucleic acid refers to a genomic region, such as an LD-block, found to be associated with risk of thyroid cancer through at least one polymorphic marker located within the region or LD block.

Variants associated with risk of thyroid cancer

The present inventors have identified genetic variation that correlates with risk of thyroid cancer. On chromosome 9q34.2, genetic variation in the RALGDS gene has been found to correlate with risk of thyroid cancer. Certain genetic variants encoding amino acid substitution in encoded RALGDS protein have been found to correlate with susceptibility of thyroid cancer. For example, allele T of rsl 13532379, encoding a Glycine to Serine substitution at position 713 in RALGDS protein, and allele C of rsl39082000, encoding a Serine to Cysteine substitution at position 686 in RALGDS, are associated with risk of thyroid cancer with OR values above 2.0. Other sequence variants in RALGDS that are also contemplated to be associated with risk of thyroid cancer are disclosed herein.

The inventors have further identified a number of variants that are associated with risk of thyroid cancer. Thus, variants on chromosome 1 (rsl2407041, rs6689698, rsl2136101 and

rsl 1586476), chromosome 2 (rs62174266 and rs76839330), chromosome 3 (rsl354833), chromosome 4 (rs75825480 and rs6828277), chromosome 7 (rsl7540362), chromosome 8 (rsl060412 and rs55635625), chromosome 9 (rs497341), chromosome 13 (rs73205431), chromosome 14 (rs8021657 and rsl46663071), chromosome 17 (rs28524987) and chromosome 21 (rs62223910) have been identified as being associated with risk of thyroid cancer. These variants, and variants in linkage disequilibrium with these variants, are all useful in the methods described in further detail herein.

Ras proteins are small monomeric GTPases that act as molecular switches by coupling extracellular signals to various cellular responses. These proteins therefore serve a critical function in the control of cellular signalling pathways that are responsible for growth, migration, adhesion, cytoskeletal integrity, survival and differentiation of cells. Ras cycles from an active to an inactive state through a mechanism that is regulated by GTP exchange factors (GEFs), which catalyze the change of GDP for GTP. Ras can interact with many effector molecules to activate parallel pathways, including c-RAF and RALGDS.

Ral guanine nucleotide dissociation stimulator (RALGDS) is one of several known Ras-regulated guanine-nucleotide exchange factors (GEFs) that lead to Ral activation, thus coupling the Ras pathway to the pathway of other GTPases. RALGDS protein contains several functional domains. A Ras-like guanine nucleotide exchange factor domain (Ras Exchange Motif; REM) is located near the N-terminus (position 110 to 250 in RALGDS protein with sequence as set forth in SEQ ID NO: 58). The protein contains a central CDC25-like GEF domain and a Ras-binding-domain (RBD)

Table of RALGDS domains. Based on Pfam and SMART protein domain predictors

Pfam

Domain Start End

low_complexity 7 18

RasGEF_N 115 227

low_complexity 279 301

low_complexity 304 323

low_complexity 325 340

RasGEF 383 597

low_complexity 675 688

low_complexity 708 717

low_complexity 745 774

RA 798 885

low_complexity 838 849

SMART

Domain Start End

low complexity 7 18

RasGEF_N 111 237

low complexity 267 289

low complexity 292 311

low complexity 313 328

RasGEF 370 637

low complexity 663 676

low complexity 696 705

low complexity 733 762

RA 786 873 at the C-terminus, which associates directly with Ras in a GTP-dependent fashion. This type of RBD has also been termed the RA (RalGDS/AF6, Ras-associating) domain. The following domain table contains a summary of functional domains in the RALGDS protein. Ras proteins are membrane-associated molecular switches that bind GTP and GDP and slowly hydrolyze GTP to GDP. The balance between the GTP bound (active) and GDP bound (inactive) states is regulated by the opposite action of proteins activating the GTPase activity and that of proteins which promote the loss of bound GDP and the uptake of fresh GTP. The latter proteins are known as guanine-nucleotide dissociation stimulators (GDSs) (or also as guanine-nucleotide releasing (or exchange) factors (GRFs)). Proteins that act as GDS can be classified into at least two families, on the basis of sequence similarities, the CDC24 family (see IPR001331) and the CDC25 family. The size of the proteins of the CDC25 family range from 309 residues (LTE1) to 1596 residues (sos). The sequence similarity shared by all these proteins is limited to a region of about 250 amino acids generally located in their C-terminal section (currently the only exceptions are sos and RALGDS where this domain makes up the central part of the protein). This domain, sometimes called RasGEF has been shown, in CDC25 and SCD25, to be essential for the activity of these proteins.

Proteins with a Ras association domain (RA), including RALGDS and AF-6 are mostly RasGTP effectors and include guanine-nucleotide releasing factor in mammals. This factor stimulates the dissociation of GDP from the Ras-related RALA and RALB GTPases, which allows GTP binding and activation of the GTPases. It interacts and acts as an effector molecule for R-ras, K-Ras and Rap. The RA domain is also present in a number of other proteins among them the sexual

differentiation protein in yeast that is essential for mating and meiosis and yeast adenylate cyclase. These proteins contain repeated leucine-rich (LRR) segments.

A subset of guanine nucleotide exchange factor for Ras-like small GTPases appear to possess a guanine nucleotide exchange factor domain N-terminal to the RasGef (Cdc25-like) domain (RasGEF_N). The recent crystal structureof sos shows that this domain is alpha-helical (Nature 394, 337-343).

Mutations in any of the above described RALGDS domains, or outside of these domains, are contemplated to affect the activity of the RALGDS polypeptide. In various embodiments of the methods of the disclosure, the residues of particular interest are those that are conserved across the species identified in Figure 4. This is because the conservation of one or more amino acids suggests an evolutionary significance, and loss or mutation of the one or more amino acids can lead to a loss in overall RALGDS activity.

The c-RAF protein is a serine/threonine-specific protein kinase that acts downstream of Ras. The protein interacts with a number of cellular partners, including B-RAF (BRAF), which is known to be mutated in human cancers. A number of mutations of the BRAF gene have been identified in human cancers. The gene is highly mutated in thyroid cancers, and is in fact the most mutated gene in thyroid tumors, representing about half of all known mutations in thyroid tumors. The Ras pathway is therefore at the forefront of the biological events in human thyroid tumors. It is therefore likely that various mutations in the human RALGDS protein are involved in thyroid cancer, including the mutations described herein.

The present inventors have identified a number of sequence variants in the human RALGDS gene that encode amino acid substitutions in RALGDS protein. These include markers rs34170541, encoding an isoleucine to leucine substitution at position 724 in RALGDS protein (SEQ ID NO: 58), marker rsl l3532379, encoding a glycine to serine substitution at position 713 in RALGDS protein (SEQ ID NO: 58), marker rsl39082000, encoding a serine to cysteine substitution at position 686 in RALGDS protein (SEQ ID NO: 58), marker chr9: 134971204, encoding a glutamine to arginine substitution at position 513 in RALGDS protein (SEQ ID NO: 58), marker rsl40573248, encoding a lycine to arginine substitution at position 222 in

RALGDS protein (SEQ ID NO: 58), and marker chr9: 134977276, encoding a glycine to cysteine substitution at position 90 in RALGDS protein (SEQ ID NO: 58).

Methods of determining susceptibility to thyroid cancer As a consequence, the present invention in one aspect provides a method of determining a susceptibility to thyroid cancer, the method comprising steps of (i) analyzing data about at least one allele of human RALGDS gene (SEQ ID NO: 57) in a human subject, wherein different alleles of the human RALGDS gene are associated with different susceptibilities to thyroid cancer in humans, and (ii) determining a susceptibility to thyroid cancer for the human subject from the data.

The data can be any type of data that is representative of polymorphic alleles in the RALGDS gene. In certain embodiments, the data is nucleic acid sequence data. The sequence data is data that is sufficient to provide information about particular alleles. In certain embodiments, the nucleic acid sequence data is obtained from a biological sample comprising or containing nucleic acid from the human individual. The nucleic acids sequence may suitably be obtained using a method that comprises at least one procedure selected from (i) amplification of nucleic acid from the biological sample; (ii) hybridization assay using a nucleic acid probe and nucleic acid from the biological sample; (iii) hybridization assay using a nucleic acid probe and nucleic acid obtained by amplification of the biological sample, and (iv) sequencing, in particular high- throughput sequencing. The nucleic acid sequence data may also be obtained from a preexisting record. For example, the preexisting record may comprise a genotype dataset for at least one polymorphic marker. In certain embodiments, the determining comprises comparing the sequence data to a database containing correlation data between the at least one polymorphic marker and susceptibility to thyroid cancer. In certain embodiments, the sequence data is provided as genotype data, identifying the presence or absence of particular alleles at polymorphic locations.

In some embodiments, the analyzing comprises analyzing the data for the presence or absence of at least one mutant allele indicative of a RALGDS defect. The RALGDS defect may for example be a missense mutation, a nonsense mutation or a premature truncation or frameshift of an encoded RALGDS protein, relative to a wild-type amino acid sequence, such as the wild- type amino acid sequence presented in SEQ ID NO: 58 herein. The RALGDS defect may also be expression of a RALGDS protein with reduced activity compared to a wild-type RALGDS protein. The activity can for example be guanine nucleotide exchange activity. The activity can also be RAS binding activity. In one embodiment, the RALGDS defect is selected from defects that impair any of these activities.

Determination of RAS binding to RALGDS or nucleotide exchange activity can be performed using standard assays well known to the skilled person . As noted above, such assays can be used to confirm that a particular RALGDS mutation impairs or eliminates a RALGDS activity and therefore would be expected to carry an increased susceptibility for thyroid cancer as described herein.

The data to be analyzed by the method of the invention is suitably obtained by analysis of a biological sample from a human subject to obtain information about particular alleles in the genome of the individual . In certain embodiments, the information is nucleic acid information which comprises sufficient sequence to identify the presence or absence of at least one allele in the subject (e.g. a mutant allele) . The information can also be nucleic acid information that identifies at least one allele of a polymorphic marker that is in linkage disequi librium with a mutant allele.

Thus, another aspect of the invention relates to a method of determining whether an individual is at increased risk of developing thyroid cancer, the method comprising steps of (i) obtaining a biological sample containing nucleic acid from the individual ; (ii) determining, in the biological sample, nucleic acid sequence about the human RALGDS gene; and (iii) comparing the sequence information to the wild-type sequence of RALGDS (SEQ ID NO: 10) ; wherein an identification of a mutation in RALGDS in the individual is indicative that the individual is at increased risk of developing thyroid cancer.

Linkage disequilibrium may suitably be determined by the correlation coefficient between polymorphic sites. In one embodiment, the sites are correlated by values of the correlation coefficient r² of greater than 0.2. In another embodiment, the sites are correlated by values of the correlation coefficient r² of greater than 0.5. Other suitable values of r² that are also appropriate to characterize polymorphic sites in LD are however also contemplated, as discussed further herein. The information may also be information a bout measurement of quantity of length of RALGDS mRNA, wherein the measurement is indicative of the presence or absence of the mutant allele. For example, mutant alleles may result in premature truncation of transcribed mRNA which can be detected by measuring the length of mRNA. The information may further be measurement of quantity of RALGDS protein, wherein the measurement of protein is indicative of the presence or absence of a mutant allele. Truncated transcripts will result in truncated forms of translated polypeptides, which can be measured using standard methods known in the art. For example, truncated proteins or proteins arising from a frameshift may have fewer or different epitopes from wildtype protein and can be distinguished with immunoassays. Truncated proteins or proteins altered in other ways may migrate differently and be distinguished with electrophoresis. The information obtained may also be measurement of RALGDS activity, wherein the measurement is indicative of the mutant allele. The activity is suitably selected from RAS binding activity and nucleotide exchange activity. In one embodiment, the information is selected from any one of the above mentioned types of information.

In a further embodiment of the invention, a biological sample is obtained from the human subject prior to the analyzing steps. The analyzing may also suitably be performed by analyzing data from a preexisting record about the human subject. The preexisting record may for example include sequence information or genotype information about the individual, which can identify the presence or absence of mutant alleles.

In certain embodiments, information about risk for the human subject can be determined using methods known in the art. Some of these methods are described herein. For example, information about odds ratio (OR), relative risk (RR) or lifetime risk (LR) can be determined from information about the presence or absence of particular mutant alleles of RAGLDS.

In certain embodiments, the allele of the human RALGDS gene is selected from alleles of the polymorphic marker rs34170541, marker rsl l3532379, marker rsl39082000, marker chr9: 134971204, marker rsl40573248, and marker chr9: 134977276. In one embodiment, the allele of the human RALGDS gene is selected from the group consisting of the G allele of the polymorphic marker rs34170541, encoding an isoleucine to leucine substitution at position 724 in RALGDS protein (SEQ ID NO: 58), the T allele of marker rsl l3532379, encoding a glycine to serine substitution at position 713 in RALGDS protein (SEQ ID NO: 58), the C allele of marker rsl39082000, encoding a serine to cysteine substitution at position 686 in RALGDS protein (SEQ ID NO: 58), the C allele of marker chr9: 134971204, encoding a glutamine to arginine substitution at position 513 in RALGDS protein (SEQ ID NO: 58), the C allele of marker rsl40573248, encoding a lycine to arginine substitution at position 222 in RALGDS protein (SEQ ID NO: 58), and the A allele of marker chr9: 134977276, encoding a glycine to cysteine substitution at position 90 in RALGDS protein (SEQ ID NO: 58).

In one preferred embodiment, the allele of the human RALGDS gene is the T allele of marker rsl l3532379. In another preferred embodiment, the allele of the human RALGDS gene is the C allele of marker rsl39082000.

As will be described in more detail in the below, the skilled person will appreciate that marker alleles in linkage disequilibrium with any one of these at-risk alleles of thyroid cancer are also predictive of increased risk of thyroid cancer, and may thus also be suitably selected for use in the methods of the invention. It may thus also be suitable to analyze sequence data for surrogate markers of particular anchor markers that have been identified as being associated with risk of thyroid cancer. In certain embodiments, suitable surrogate markers are markers that are correlated to an anchor marker by values of r² of at least 0.2. In certain embodiments, suitable surrogate markers are correlated to the anchor marker by values of r² of at least 0.5.

Certain alleles of risk variants of thyroid cancer are predictive of increased risk (increased susceptibility) of thyroid cancer. Thus, the G allele of rs34170541, the T allele of rsl 13532379, the C allele of rsl39082000, the C allele of chr9: 134971204, the C allele of rsl40573248, and the A allele of chr9: 134977276, are predictive of increased risk of thyroid cancer. Thus, in certain embodiment, determination of the presence of any one of these alleles is indicative of increased risk of thyroid cancer for the individual. Determination of the absence of any of these alleles is indicative that the individual does not have the increased risk conferred by the allele. In other words, the individual is at a decreased risk of thyroid cancer compared with individuals who carry at least one copy of the allele in their genome.

The allele that is detected can suitably be the allele of the complementary strand of DNA, such that the nucleic acid sequence data includes the identification of at least one allele which is complementary to any of the alleles of the polymorphic markers referenced above. For example, the allele that is detected may be the complementary A allele of the at-risk T allele of rsl l3532379.

Another aspect of the invention relates to a method for determining a susceptibility to thyroid cancer in a human individual, comprising determining the presence or absence of at least one allele of at least one polymorphic marker in a nucleic acid sample obtained from the individual, wherein the at least one polymorphic marker is selected from the group consisting of (a) polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58; and (b) rsl2407041, rs6689698, rsl2136101, rsl l586476, rs62174266, rs76839330, rsl354833, rs75825480, rs6828277, rsl7540362, rsl060412, rs55635625, rs497341, rs73205431, rs8021657, rsl46663071, rs28524987 and rs62223910; and

determining a susceptibility to thyroid cancer from the presence or absence of the at least one allele, wherein the presence of the at least one allele is indicative of a susceptibility to thyroid cancer.

In one embodiment, the polymorphic marker is a polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58. In another embodiment, the polymorphic marker is selected from the group consisting of rsl2407041, rs6689698, rsl2136101, rsl 1586476, rs62174266, rs76839330, rsl354833, rs75825480, rs6828277, rsl7540362, rsl060412, rs55635625, rs497341, rs73205431, rs8021657, rsl46663071, rs28524987 and rs62223910. In another embodiment, the polymorphic marker is a correlated marker in linkage disequilibrium with at least one marker selected from the group consisting of rsl2407041, rs6689698, rsl2136101, rsl 1586476, rs62174266, rs76839330, rsl354833, rs75825480, rs6828277, rsl7540362, rsl060412, rs55635625, rs497341, rs73205431, rs8021657, rsl46663071, rs28524987 and rs62223910.

In another aspect, a method is provided that comprises (1) obtaining a sample containing nucleic acid from a human individual; (2) obtaining nucleic acid sequence data about at least one polymorphic marker in the sample, wherein different alleles of the at least one marker are associated with different susceptibilities of thyroid cancer in humans; (3) analyzing the nucleic acid sequence data about the at least one marker; and (4) determining a risk of thyroid cancer from the nucleic acid sequence data. In certain embodiments, the analyzing comprises determining the presence or absence of at least one allele of the at least one polymorphic marker. In one embodiment, the polymorphic marker is a polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58.

It is contemplated that in certain embodiments of the invention, it may be convenient to prepare a report of results of risk assessment. Thus, certain embodiments of the methods of the invention comprise a further step of preparing a report containing results from the

determination, wherein said report is written in a computer readable medium, printed on paper, or displayed on a visual display. In certain embodiments, it may be convenient to report results of susceptibility to at least one entity selected from the group consisting of the individual, a guardian of the individual, a genetic service provider, a physician, a medical organization, and a medical insurer.

In another aspect, the invention relates to a method of determining a susceptibility to thyroid cancer in a human individual, comprising determining whether at least one at-risk allele in at least one polymorphic marker is present in a genotype dataset derived from the individual, wherein the at least one polymorphic marker is a polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58, and wherein determination of the presence of the at least one at-risk allele is indicative of increased susceptibility to thyroid cancer in the individual.

A genotype dataset derived from an individual is in the present context a collection of genotype data that is indicative of the genetic status of the individual for particular genetic markers. The dataset is derived from the individual in the sense that the dataset has been generated using genetic material from the individual, or by other methods available for determining genotypes at particular genetic markers (e.g., imputation methods). The genotype dataset comprises in one embodiment information about marker identity and the allelic status of the individual for at least one allele of a marker, i.e. information about the identity of at least one allele of the marker in the individual. The genotype dataset may comprise allelic information (information about allelic status) about one or more marker, including two or more markers, three or more markers, five or more markers, ten or more markers, one hundred or more markers, and so on. In some embodiments, the genotype dataset comprises genotype information from a whole-genome assessment of the individual, which may include hundreds of thousands of markers, or even one million or more markers spanning the entire genome of the individual.

In any of the methods described herein, the human subject or human individual whose susceptibility of thyroid cancer is being assessed may be a male or a female. In certain embodiments, the human subject is a female.

In certain embodiments, the methods of the invention relate to determination of susceptibility of thyroid cancer with an early onset. In one such embodiment, the methods relate to

determination of susceptibility of thyroid cancer with an early onset in female subjects. In one embodiment, the methods relate to thyroid cancer with an onset before age 60 years. In another embodiment, the methods relate to thyroid cancer with an onset before age 50 years. in a further embodiment, the methods relate to thyroid cancer with an onset before age 40 years.

Diagnostic assays for thyroid cancer

The present invention also provides assays that are useful for determining susceptibility of thyroid cancer in humans. One such assay comprises steps of (i) obtaining a nucleic acid sample from the human subject; (ii) assaying the nucleic acid sample to determine the presence or absence of at least one allele of at least one polymorphic marker associated with increased susceptibility to thyroid cancer in humans, and (iii) determining a susceptibility to thyroid cancer for the human subject from the presence or absence of the at least one allele.

In one embodiment the at least one polymorphic marker is selected from the group consisting of (a) polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58; and (b) rsl2407041, rs6689698, rsl2136101, rsl l586476, rs62174266, rs76839330, rsl354833, rs75825480, rs6828277, rsl7540362, rsl060412, rs55635625, rs497341, rs73205431, rs8021657, rsl46663071, rs28524987 and rs62223910; and wherein a susceptibility to thyroid cancer is determined from the presence or absence of the at least one allele. In certain embodiments, determination of susceptibility is made based on the presence or absence of the at least one allele. Determination of the presence of at least one allele conferring increased risk of thyroid cancer (at-risk allele) is indicative that the human subject is at increased susceptibility of thyroid cancer. Determination of the absence of at least one such allele is indicative that the individual does not have the elevated susceptibility. In one embodiment, the subject is at decreased susceptibility of thyroid cancer compared with subjects who carry at least one copy of the at-risk allele in their genome.

Assessment of other biomarkers for thyroid cancer

Certain embodiments of the invention further comprise assessing the quantitative levels of a biomarker for thyroid cancer. For example, the levels of a biomarker may be determined in concert with analysis of particular genetic markers. Alternatively, biomarker levels are determined at a different point in time, but results of such determination are used together with results from sequencing analysis for particular polymorphic markers. The biomarker may in some embodiments be assessed in a biological sample from the individual. In some

embodiments, the sample is a blood sample. The blood sample is in some embodiments a serum sample. In preferred embodiments, the biomarker is selected from the group consisting of thyroid stimulating hormone (TSH), thyroxine (T4) and thriiodothyronine (T3). In certain embodiments, determination of an abnormal level of the biomarker is indicative of an abnormal thyroid function in the individual, which may in turn be indicative of an increased risk of thyroid cancer in the individual. The abnormal level can be an increased level or the abnormal level can be a decreased level. In certain embodiments, the determination of an abnormal level is determined based on determination of a deviation from the average levels of the biomarker in the population. In one embodiment, abnormal levels of TSH are measurements of less than 0.2mIU/L and/or greater than lOmlU/L In another embodiment, abnormal levels of TSH are measurements of less than 0.3mIU/L and/or greater than 3.0mIU/L In another embodiment, abnormal levels of T₃ (free T₃) are less than 70 ng/dL and/or greater than 205 ng/dL In another embodiment, abnormal levels of T₄ (free T₄) are less than 0.8 ng/dL and/or greater than 2.7 ng/dL.

The markers conferring risk of thyroid cancer, as described herein, can be combined with other genetic markers for thyroid cancer. Such markers are typically not in linkage disequilibrium with rsl 13532379, or other markers described herein to be predictive of risk of thyroid cancer. Any of the methods described herein can be practiced by combining the genetic risk factors described herein with additional genetic risk factors for thyroid cancer.

Thus, in certain embodiments, a further step is included, comprising determining whether at least one at-risk allele of at least one at-risk variant for thyroid cancer not in linkage

disequilibrium with rsl l3532379 is present in a sample comprising genomic DNA from a human individual or a genotype dataset derived from a human individual. In other words, genetic markers in other locations in the genome can be useful in combination with the marker of the present invention, so as to determine overall risk of thyroid cancer based on multiple genetic variants. Selection of markers that are not in linkage disequilibrium (not in LD) can be based on a suitable measure for linkage disequilibrium, as described further herein. In certain

embodiments, markers that are not in linkage disequilibrium have values of the LD measure r² correlating the markers of less than 0.2. In certain other embodiments, markers that are not in LD have values for r² correlating the markers of less than 0.15, including less than 0.10, less than 0.05, less than 0.02 and less than 0.01. Other suitable numerical values for establishing that markers are not in LD are contemplated, including values bridging any of the above- mentioned values.

In one embodiment, assessment of the marker described herein is combined with assessment of at least one marker selected from the group consisting of marker rs965513 on chromosome 9q22, marker rs944289 on chromosome 14q l3, marker rs7005606 on chromosome 8pl2 and marker rs966423 on chromosome 2q35, or a marker in linkage disequilibrium therewith, to establish overall risk. In certain such embodiments, determination of the presence of the A allele of rs965513, the T allele of rs944289, the G allele of rs7005606 and/or the C allele of rs966423 is indicative of increased risk of thyroid cancer. In one embodiment, the A allele of rs965513 is an at-risk allele of thyroid cancer, the T allele of rs944289 is an at-risk allele of thyroid cancer, the G allele of rs7005606 is an at-risk allele of thyroid cancer and the C allele of rs966423 is an at-risk allele of thyroid cancer.

In certain embodiments, multiple markers as described herein are determined to determine overall risk of thyroid cancer. Thus, in certain embodiments, an additional step is included, the step comprising determining whether at least one allele in each of at least two polymorphic markers is present in a sample comprising genomic DNA from a human individual or a genotype dataset derived from a human individual, wherein the presence of the at least one allele in the at least two polymorphic markers is indicative of an increased susceptibility to thyroid cancer. The genetic markers of the invention can also be combined with non-genetic information to establish overall risk for an individual. Thus, in certain embodiments, a further step is included, comprising analyzing non-genetic information to make risk assessment, diagnosis, or prognosis of the individual. The non-genetic information can be any information pertaining to the disease status of the individual or other information that can influence the estimate of overall risk of thyroid cancer for the individual. In one embodiment, the non-genetic information is selected from age, gender, ethnicity, socioeconomic status, previous disease diagnosis, medical history of subject, family history of thyroid cancer, biochemical measurements, and clinical measurements.

Obtaining nucleic acid sequence data

Sequence data can be nucleic acid sequence data, which may be obtained by means known in the art. Sequence data is suitably obtained from a biological sample of genomic DNA, RNA, or cDNA (a "test sample") from an individual ("test subject). For example, nucleic acid sequence data may be obtained through direct analysis of the sequence of the polymorphic position (allele) of a polymorphic marker. Suitable methods, some of which are described herein, include, for instance, whole genome sequencing methods, whole genome analysis using SNP chips (e.g., Infinium HD BeadChip), cloning for polymorphisms, non-radioactive PCR-single strand conformation polymorphism analysis, denaturing high pressure liquid chromatography (DHPLC), DNA hybridization, computational analysis, single-stranded conformational polymorphism

(SSCP), restriction fragment length polymorphism (RFLP), automated fluorescent sequencing; clamped denaturing gel electrophoresis (CDGE); denaturing gradient gel electrophoresis (DGGE), mobility shift analysis, restriction enzyme analysis; heteroduplex analysis, chemical mismatch cleavage (CMC), RNase protection assays, use of polypeptides that recognize nucleotide mismatches, such as E. coli mutS protein, allele-specific PCR, and direct manual and automated sequencing. These and other methods are described in the art (see, for instance, Li et al., Nucleic Acids Research, 28(2) : el (i-v) (2000); Liu et al., Biochem Cell Bio 80: 17-22 (2000); and Burczak et al., Polymorphism Detection and Analysis, Eaton Publishing, 2000; Sheffield et al., Proc. Natl. Acad. Sci. USA, 86:232-236 (1989); Orita et al., Proc. Natl. Acad. Sci. USA, 86:2766-2770 (1989); Flavell et al., Cell, 15:25-41 (1978); Geever et al., Proc. Natl. Acad. Sci. USA, 78: 5081-5085 (1981); Cotton et al., Proc. Natl. Acad. Sci. USA, 85:4397-4401 (1985); Myers et al., Science 230: 1242-1246 (1985); Church and Gilbert, Proc. Natl. Acad. Sci. USA, 81 : 1991-1995 (1988); Sanger et al., Proc. Natl. Acad. Sci. USA, 74: 5463-5467 (1977); and Beavis et al., U.S. Patent No. 5,288,644).

Recent technological advances have resulted in technologies that allow massive parallel sequencing to be performed in relatively condensed format. These technologies share sequencing-by-synthesis principle for generating sequence information, with different technological solutions implemented for extending, tagging and detecting sequences. Exemplary technologies include 454 pyrosequencing technology (Nyren, P. et al. Anal Biochem 208: 171-75 (1993); http://www.454.com), Illumina Solexa sequencing technology (Bentley, D.R. Curr Opin Genet Dev 16: 545-52 (2006); http://www.illumina.com), and the SOLiD technology developed by Applied Biosystems (ABI) (http://www.appliedbiosystems.com; see also Strausberg, R.L., et al. Drug Disc Today 13: 569-77 (2008)). Other sequencing technologies include those developed by Pacific Biosciences (http://www.pacificbiosciences.com), Complete Genomics

(http://www.completegenomics.com), Intelligen Bio-Systems

(http://www.intelligentbiosystems.com), Genome Corp (http://www.genomecorp.com), ION Torrent Systems (http://www.iontorrent.com) and Helicos Biosciences

(http://www.helicosbio.som). It is contemplated that sequence data useful for performing the present invention may be obtained by any such sequencing method, or other sequencing methods that are developed or made available. Thus, any sequence method that provides the allelic identity at particular polymorphic sites (e.g., the absence or presence of particular alleles at particular polymorphic sites) is useful in the methods described and claimed herein.

Alternatively, hybridization methods may be used (see Current Protocols in Molecular Biology, Ausubel et al., eds., John Wiley & Sons, including all supplements). For example, a biological sample of genomic DNA, RNA, or cDNA (a "test sample") may be obtained from a test subject. The subject can be an adult, child, or fetus. The DNA, RNA, or cDNA sample is then examined. The presence of a specific marker allele can be indicated by sequence-specific hybridization of a nucleic acid probe specific for the particular allele. The presence of more than one specific marker allele or a specific haplotype can be indicated by using several sequence-specific nucleic acid probes, each being specific for a particular allele. A sequence-specific probe can be directed to hybridize to genomic DNA, RNA, or cDNA. A "nucleic acid probe", as used herein, can be a DNA probe or an RNA probe that hybridizes to a complementary sequence. One of skill in the art would know how to design such a probe so that sequence specific hybridization will occur only if a particular allele is present in a genomic sequence from a test sample.

In certain embodiments, determination of a susceptibility to thyroid cancer comprises forming a hybridization sample by contacting the test sample, such as a genomic DNA sample, with at least one nucleic acid probe. A non-limiting example of a probe for detecting mRNA or genomic DNA is a labeled nucleic acid probe that is capable of hybridizing to mRNA or genomic DNA sequences described herein. The nucleic acid probe can be, for example, a full-length nucleic acid molecule, or a portion thereof, such as an oligonucleotide of at least 10, 15, 30, 50, 100, 250 or 500 nucleotides in length that is sufficient to specifically hybridize under stringent conditions to appropriate mRNA or genomic DNA. For example, the nucleic acid probe can comprise all or a portion of the nucleotide sequence of the RALGDS gene, or the probe can be the complementary sequence of such a sequence. Hybridization can be performed by methods well known to the person skilled in the art (see, e.g., Current Protocols in Molecular Biology, Ausubel et al., eds., John Wiley & Sons, including all supplements). In one embodiment, hybridization refers to specific hybridization, i.e., hybridization with no mismatches (exact hybridization). In one embodiment, the hybridization conditions for specific hybridization are high stringency.

Specific hybridization, if present, is detected using standard methods. If specific hybridization occurs between the nucleic acid probe and the nucleic acid in the test sample, then the sample contains the allele that is complementary to the nucleotide that is present in the nucleic acid probe. Additionally, or alternatively, a peptide nucleic acid (PNA) probe can be used in addition to, or instead of, a nucleic acid probe in the hybridization methods described herein. A PNA is a DNA mimic having a peptide-like, inorganic backbone, such as N-(2-aminoethyl)glycine units, with an organic base (A, G, C, T or U) attached to the glycine nitrogen via a methylene carbonyl linker (see, for example, Nielsen et al., Bioconjug. Chem. 5:3-7 (1994)). The PNA probe can be designed to specifically hybridize to a molecule in a sample suspected of containing one or more of the marker alleles that are associated with risk of thyroid cancer.

In one embodiment of the invention, a test sample containing genomic DNA obtained from the subject is collected and the polymerase chain reaction (PCR) is used to amplify a fragment comprising one or more polymorphic marker. As described herein, identification of particular marker alleles can be accomplished using a variety of methods. In another embodiment, determination of a susceptibility is accomplished by expression analysis, for example using quantitative PCR (kinetic thermal cycling). This technique can, for example, utilize commercially available technologies, such as TaqMan® (Applied Biosystems, Foster City, CA) . The technique can for example assess the presence of an alteration in the expression or composition of a polypeptide or splicing variant(s) that is encoded by a nucleic acid associated described herein. Alternatively, this technique may assess expression levels of genes or particular splice variants of genes, that are affected by one or more of the variants described herein. Further, the expression of the variant(s) can be quantified as physically or functionally different.

Allele-specific oligonucleotides can also be used to detect the presence of a particular allele in a nucleic acid. An "allele-specific oligonucleotide" (also referred to herein as an "allele-specific oligonucleotide probe") is an oligonucleotide of any suitable size, for example an oligonucleotide of approximately 10-50 base pairs or approximately 15-30 base pairs, that specifically hybridizes to a nucleic acid which contains a specific allele at a polymorphic site (e.g., a polymorphic marker). An allele-specific oligonucleotide probe that is specific for one or more particular alleles at polymorphic markers can be prepared using standard methods (see, e.g., Current Protocols in Molecular Biology, supra). PCR can be used to amplify the desired region. Specific hybridization of an allele-specific oligonucleotide probe to DNA from a subject is indicative of the presence of a specific allele at a polymorphic site (see, e.g., Gibbs et al., Nucleic Acids Res. 17:2437-2448 (1989) and WO 93/22456).

With the addition of analogs such as locked nucleic acids (LNAs), the size of primers and probes can be reduced to as few as 8 bases. LNAs are a novel class of bicyclic DNA analogs in which the 2' and 4' positions in the furanose ring are joined via an O-methylene (oxy-LNA), S-methylene (thio-LNA), or amino methylene (amino-LNA) moiety. Common to all of these LNA variants is an affinity toward complementary nucleic acids, which is by far the highest reported for a DNA analog. For example, particular all oxy-LNA nonamers have been shown to have melting temperatures (Tm) of 64°C and 74°C when in complex with complementary DNA or RNA, respectively, as opposed to 28°C for both DNA and RNA for the corresponding DNA nonamer. Substantial increases in Tm are also obtained when LNA monomers are used in combination with standard DNA or RNA monomers. For primers and probes, depending on where the LNA monomers are included (e.g., the 3' end, the 5' end, or in the middle), the Tm could be increased considerably. It is therefore contemplated that in certain embodiments, LNAs are used to detect particular alleles at polymorphic sites associated with particular vascular conditions, as described herein.

In certain embodiments, arrays of oligonucleotide probes that are complementary to target nucleic acid sequence segments from a subject can be used to identify polymorphisms in a nucleic acid. For example, an oligonucleotide array can be used. Oligonucleotide arrays typically comprise a plurality of different oligonucleotide probes that are coupled to a surface of a substrate in different known locations. These arrays can generally be produced using mechanical synthesis methods or light directed synthesis methods that incorporate a combination of photolithographic methods and solid phase oligonucleotide synthesis methods, or by other methods known to the person skilled in the art (see, e.g., Bier et al., Adv Biochem Eng

Biotechnol 109:433-53 (2008); Hoheisel, Nat Rev Genet 7:200-10 (2006); Fan et al., Methods Enzymol 410: 57-73 (2006); Raqoussis & Elvidge, Expert Rev Mol Diagn 6: 145-52 (2006);

Mockler et al., Genomics 85: 1-15 (2005), and references cited therein, the entire teachings of each of which are incorporated by reference herein). Many additional descriptions of the preparation and use of oligonucleotide arrays for detection of polymorphisms can be found, for example, in US 6,858,394, US 6,429,027, US 5,445,934, US 5,700,637, US 5,744,305, US 5,945,334, US 6,054,270, US 6,300,063, US 6,733,977, US 7,364,858, EP 619 321, and EP 373 203, the entire teachings of which are incorporated by reference herein.

Also, standard techniques for genotyping can be used to detect particular marker alleles, such as fluorescence-based techniques (e.g., Chen et al., Genome Res. 9(5) : 492-98 (1999); Kutyavin et al., Nucleic Acid Res. 34:el28 (2006)), utilizing PCR, LCR, Nested PCR and other techniques for nucleic acid amplification. Specific commercial methodologies available for SNP genotyping include, but are not limited to, TaqMan genotyping assays and SNPlex platforms (Applied

Biosystems), gel electrophoresis (Applied Biosystems), mass spectrometry (e.g., MassARRAY system from Sequenom), minisequencing methods, real-time PCR, Bio-Plex system (BioRad), CEQ and SNPstream systems (Beckman), array hybridization technology(e.g., Affymetrix

GeneChip; Perlegen ), BeadArray Technologies (e.g., Illumina GoldenGate and Infinium assays), array tag technology (e.g., Parallele), and endonuclease-based fluorescence hybridization technology (Invader; Third Wave).

Suitable biological sample in the methods described herein can be any sample containing nucleic acid (e.g., genomic DNA) and/or protein from the human individual. For example, the biological sample can be a blood sample, a serum sample, a leukapheresis sample, an amniotic fluid sample, a cerbrospinal fluid sample, a hair sample, a tissue sample from skin, muscle, buccal, or conjuctival mucosa, placenta, gastrointestinal tract, or other organs, a semen sample, a urine sample, a saliva sample, a nail sample, a tooth sample, and the like. Preferably, the sample is a blood sample, a salive sample or a buccal swab.

Protein analysis

Missense nucleic acid variations may lead to an altered amino acid sequence, as compared to the non-variant (e.g., wild-type) protein, due to one or more amino acid substitutions, deletions, or insertions, or truncation (due to, e.g., splice variation). In such instances, detection of the amino acid substitution of the variant protein may be useful. This way, nucleic acid sequence data may be obtained through indirect analysis of the nucleic acid sequence of the allele of the polymorphic marker, i.e. by detecting a protein variation. Methods of detecting variant proteins are known in the art. For example, direct amino acid sequencing of the variant protein followed by comparison to a reference amino acid sequence can be used. Alternatively, SDS-PAGE followed by gel staining can be used to detect variant proteins of different molecular weights. Also, Immunoassays, e.g., immunofluorescent immunoassays, immunoprecipitations, radioimmunoasays, ELISA, and Western blotting, in which an antibody specific for an epitope comprising the variant sequence among the variant protein and non-variant or wild-type protein can be used. In certain embodiments of the present invention, RALGDS with an altered amino acid composition compared with wild-type RALGDS is detected. In certain embodiments, RALGDS containing one or sequence variations is detected. In certain embodiments, a sequence variation selected from G713S substitution, I724L substitution, S686C substitution, Q513R substitution, K222R substitution and G90C substitution in RALGDS (SEQ ID NO: 58) is detected in a protein sample. The detection may be suitably performed using any of the methods described in the herein.

In some cases, a variant protein has altered (e.g., upregulated or downregulated) biological activity, in comparison to the non-variant or wild-type protein. The biological activity can be, for example, a binding activity or enzymatic activity. In this instance, altered biological activity may be used to detect a variation in protein encoded by a nucleic acid sequence variation. Methods of detecting binding activity and enzymatic activity are known in the art and include, for instance, ELISA, competitive binding assays, quantitative binding assays using instruments such as, for example, a Biacore® 3000 instrument, chromatographic assays, e.g., HPLC and TLC. Alternatively or additionally, a protein variation encoded by a genetic variation could lead to an altered expression level, e.g., an increased expression level of an mRNA or protein, a decreased expression level of an mRNA or protein. In such instances, nucleic acid sequence data about the allele of the polymorphic marker, or protein sequence data about the protein variation, can be obtained through detection of the altered expression level. Methods of detecting expression levels are known in the art. For example, ELISA, radioimmunoassays, immunofluorescence, and Western blotting can be used to compare the expression of protein levels. Alternatively, Northern blotting can be used to compare the levels of mRNA. These processes are described in Sambrook et al., Molecular Cloning: A Laboratory Manual, 3^rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York (2001).

Any of these methods may be performed using a nucleic acid (e.g., DNA, mRNA) or protein of a biological sample obtained from the human individual for whom a susceptibility is being determined. The biological sample can be any nucleic acid or protein containing sample obtained from the human individual. For example, the biological sample can be any of the biological samples described herein. It is further contemplated that additional missense variants in human RALGDS protein may be association with thyroid cancer risk. The present invention thus also encompasses methods of determining susceptibility of thyroid cancer, using further missense variants in human RALGDS that confer risk of thyroid cancer. Number of Polymorphic Markers/Genes Analyzed

With regard to the methods of determining a susceptibility described herein, the methods can comprise obtaining sequence data about any number of polymorphic markers and/or about any number of genes. For example, the method can comprise obtaining sequence data for about at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 40, 50, 100, 500, 1000, 10,000 or more polymorphic markers. In certain embodiments, the sequence data is obtained from a microarray comprising probes for detecting a plurality of markers. For example, the markers can be independent of markers in the RALGDS gene, or the markers may be independent of such markers. The polymorphic markers can be the ones of the group specified herein or they can be different polymorphic markers that are not listed herein. In a specific embodiment, the method comprises obtaining sequence data about at least two polymorphic markers. In certain embodiments, each of the markers may be associated with a different gene. For example, in some instances, if the method comprises obtaining nucleic acid data about a human individual identifying at least one allele of a polymorphic marker, then the method comprises identifying at least one allele of at least one polymorphic marker. Also, for example, the method can comprise obtaining sequence data about a human individual identifying alleles of multiple, independent markers, which are not in linkage disequilibrium.

Linkage Disequilibrium

Linkage Disequilibrium (LD) refers to a non-random assortment of two genetic elements. For example, if a particular genetic element (e.g. , an allele of a polymorphic marker, or a haplotype) occurs in a population at a frequency of 0.50 (50%) and another element occurs at a frequency of 0.50 (50%), then the predicted occurrance of a person's having both elements is 0.25 (25%), assuming a random distribution of the elements. However, if it is discovered that the two elements occur together at a frequency higher than 0.25, then the elements are said to be in linkage disequilibrium, since they tend to be inherited together at a higher rate than what their independent frequencies of occurrence (e.g., allele or haplotype frequencies) would predict. Roughly speaking, LD is generally correlated with the frequency of recombination events between the two elements. Allele or haplotype frequencies can be determined in a population by genotyping individuals in a population and determining the frequency of the occurence of each allele or haplotype in the population. For populations of diploids, e.g., human populations, individuals will typically have two alleles for each genetic element (e.g., a marker, haplotype or gene).

Many different measures have been proposed for assessing the strength of linkage disequilibrium (LD; reviewed in Devlin, B. & Risch, N., Genomics 29:311-22 (1995)). Most capture the strength of association between pairs of biallelic sites. Two important pairwise measures of LD are r² (sometimes denoted Δ²) and |D'| (Lewontin, R., Genetics 49:49-67 (1964); Hill, W.G. &

Robertson, A. Theor. Appl. Genet 22:226-231 (1968)) . Both measures range from 0 (no disequilibrium) to 1 ('complete' disequilibrium), but their interpretation is slightly different. |D'| is defined in such a way that it is equal to 1 if just two or three of the possible haplotypes are present, and it is < 1 if all four possible haplotypes are present. Therefore, a value of |D'| that is < 1 indicates that historical recombination may have occurred between two sites (recurrent mutation can also cause | D'| to be < 1, but for single nucleotide polymorphisms (SNPs) this is usually regarded as being less likely than recombination). The correlation measure r² represents the statistical correlation between two sites, and takes the value of 1 if only two haplotypes are present.

The r² measure is arguably the most relevant measure for association mapping, because there is a simple inverse relationship between r² and the sample size required to detect association between susceptibility loci and SNPs. These measures are defined for pairs of sites, but for some applications a determination of how strong LD is across an entire region that contains many polymorphic sites might be desirable (e.g., testing whether the strength of LD differs significantly among loci or across populations, or whether there is more or less LD in a region than predicted under a particular model). Measuring LD across a region is not straightforward, but one approach is to use the measure r, which was developed in population genetics. Roughly speaking, r measures how much recombination would be required under a particular population model to generate the LD that is seen in the data. This type of method can potentially also provide a statistically rigorous approach to the problem of determining whether LD data provide evidence for the presence of recombination hotspots.

For the methods described herein, a significant r² value can be at least 0.1 such as at least 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5, 0.55, 0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.91, 0.92, 0.93, 0.94, 0.95, 0.96, 0.97, 0.98, 0.99 or 1.0. In one specific embodiment of invention, the significant r² value can be at least 0.2. In another specific embodiment of invention, the significant r² value can be at least 0.5. In one specific embodiment of invention, the significant r² value can be at least 0.8. Alternatively, linkage disequilibrium as described herein, refers to linkage disequilibrium characterized by values of r² of at least 0.2, such as 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.85, 0.9, 0.95, 0.96, 0.97, 0.98, 0.99. Thus, linkage disequilibrium represents a correlation between alleles of distinct markers. It is measured by correlation coefficient or | D'| (r² up to 1.0 and |D'| up to 1.0). Linkage disequilibrium can be determined in a single human population, as defined herein, or it can be determined in a collection of samples comprising individuals from more than one human population. In one embodiment of the invention, LD is determined in a sample from one or more of the HapMap populations. These include samples from the Yoruba people of Ibadan, Nigeria (YRI), samples from individuals from the Tokyo area in Japan (JPT), samples from individuals Beijing, China (CHB), and samples from U.S. residents with northern and western European ancestry (CEU), as described (The International HapMap Consortium, Nature 426:789-796 (2003)). In one such embodiment, LD is determined in the Caucasian CEU population of the HapMap samples. In another embodiment, LD is determined in the African YRI population. In yet another embodiment, LD is determined in samples from the Icelandic population.

If all polymorphisms in the genome were independent at the population level (i.e., no LD between polymorphisms), then every single one of them would need to be investigated in association studies, to assess all different polymorphic states. However, due to linkage disequilibrium between polymorphisms, tightly linked polymorphisms are strongly correlated, which reduces the number of polymorphisms that need to be investigated in an association study to observe a significant association. Another consequence of LD is that many polymorphisms may give an association signal due to the fact that these polymorphisms are strongly correlated. Genomic LD maps have been generated across the genome, and such LD maps have been proposed to serve as framework for mapping disease-genes (Risch, N. & Merkiangas, K, Science 273: 1516-1517 (1996); Maniatis, N., et al., Proc Natl Acad Sci USA 99:2228-2233 (2002); Reich, DE et al, Nature 411 : 199-204 (2001)).

It is now established that many portions of the human genome can be broken into series of discrete haplotype blocks containing a few common haplotypes; for these blocks, linkage disequilibrium data provides little evidence indicating recombination (see, e.g., Wall., J.D. and Pritchard, J.K., Nature Reviews Genetics 4: 587-597 (2003); Daly, M. et al., Nature Genet.

29:229-232 (2001); Gabriel, S.B. et al., Science 296:2225-2229 (2002); Patil, N. et al., Science 294: 1719-1723 (2001); Dawson, E. et al., Nature 418:544-548 (2002); Phillips, M.S. et al., Nature Genet. 33:382-387 (2003)).

Haplotype blocks (LD blocks) can be used to map associations between phenotype and haplotype status, using single markers or haplotypes comprising a plurality of markers. The main haplotypes can be identified in each haplotype block, and then a set of "tagging" SNPs or markers (the smallest set of SNPs or markers needed to disting uish among the haplotypes) can then be identified. These tagging SNPs or markers can then be used in assessment of samples from groups of individuals, in order to identify association between phenotype and haplotype. If desired, neighboring haplotype blocks can be assessed concurrently, as there may also exist linkage disequilibrium among the haplotype blocks.

It has thus become apparent that for any given observed association to a polymorphic marker in the genome, it is likely that additional markers i n the genome also show association. This is a natural consequence of the uneven distribution of LD across the genome, as observed by the large variation in recombination rates. The markers used to detect association thus in a sense represent "tags" for a genomic region (i.e., a haplotype block or LD block) that is associating with a given disease or trait, and as such are useful for use in the methods and kits of the invention.

By way of example, the marker rsl 13532379 may be detected directly to determine risk of Thyroid Cancer. Alternatively, any marker in linkage disequilibrium with rsl l3532379, in particular markers that are closely correlated with rsl 13532379, may be detected to determine risk. The present invention thus refers to particular genetic markers for detecting association to Thyroid Cancer, as well as markers in linkage disequilibrium with these markers. Thus, in certain embodiments of the invention, markers that are in LD with this marker, e.g., markers as described herein, may be used as surrogate markers.

Suitable surrogate markers may be selected using public information, such as from the

International HapMap Consortium (http://www.hapmap.org) and the International lOOOgenomes Consortium (http://www.1000genomes.org). Publically available software may be used to identify suitable surrogate markers, for example markers that fulfill selected criteria of the LD measures r² and D'. One such software tool is available through the Broad Institute

(http://www.broadinstitute.org/mpg/snap/ldsearch.php). The stronger the linkage

disequilibrium, in particular in terms of the correlation coefficient r², to the anchor marker, the better the surrogate, and thus the mores similar the association detected by the surrogate is expected to be to the association detected by the anchor marker. Markers with values of r² equal to 1 are perfect surrogates for the at-risk variants, i.e. genotypes for one marker perfectly predicts genotypes for the other. In other words, the surrogate will, by necessity, give exactly the same association data to any particular disease as the anchor marker. Markers with smaller values of r² than 1 can also be surrogates for the at-risk anchor variant.

The present invention encompasses the assessment of such surrogate markers for the markers as disclosed herein. Such markers are annotated, mapped and listed in public databases, as well known to the skilled person, or can alternatively be readily identified by sequencing the region or a part of the region identified by the markers of the present invention in a group of individuals, and identify polymorphisms in the resulting group of sequences. As a consequence, the person skilled in the art can readily and without undue experimentation identify and select appropriate surrogate markers.

In certain embodiments, suitable surrogate markers of rsl 13532379 are selected from the group consisting of the markers set forth in Table 1.

Table 1. Surrogate markers on Chromosome 9 that are correlated by values of the correlation coefficient r²>0.2 to anchor marker rsl 13532379. Correlation data was obtained from whole-genome sequencing data of over 700 Icelandic individuals. Shown is: Marker identity, rs-name in dbSNP where available, position of that marker in NCBI Build 36 and Build 37, values of r², the predicted risk allele that is correlated with risk allele of the anchor- marker (allele T of rsl l3532379), the other allele and finally sequence ID number of flanking sequence for the marker. Allelic codes are A = 1, C = 2, G = 3, T = 4.

dbSNP-134 Build36 Build37 Risk Other Seq

Marker r²

name position position allele allele ID chr9 133971560 133971560 134981738 0,205003 A G 1 chr9 134086681 134086681 135096859 0,20567 A G 2 chr9 134315428 134315428 135325606 0,263632 G A 3 chr9 134331465 134331465 135341643 0,294956 T C 4 chr9 134434167 134434167 135444345 0,203928 A G 5 chr9 134437326 134437326 135447504 0,205957 A G 6 chr9 134511331 rsll3491378 134511331 135521509 0,265601 A G 7 chr9 134513911 134513911 135524089 0,418413 C G 8 chr9 134643499 134643499 135653677 0,509938 T C 9 chr9 134699016 rsll2376985 134699016 135709194 0,20435 A G 10 chr9 134753153 rsll3435170 134753153 135763331 0,672857 A G 11 chr9 134762502 rs45468995 134762502 135772680 0,595266 A G 12 chr9 134841310 rsl38882580 134841310 135851488 0,248776 T C 13 chr9 134841311 rsl49437446 134841311 135851489 0,248776 T A 14 chr9 134850793 rsl l l796371 134850793 135860971 0,315905 A G 15 chr9 134880456 rsl l l829392 134880456 135890634 0,217962 C T 16 chr9 134887297 rsl 12490390 134887297 135897475 0,219529 C T 17 chr9 134917247 134917247 135927425 0,815905 T c 18 chr9 134923027 rsl l l893665 134923027 135933205 0,530663 c T 19 chr9 134967220 rsll3532379 134967220 135977398 1 T c 20 chr9 135042093 rsl l2874295 135042093 136052271 0,250992 T c 21 chr9 135285914 rsl l l238709 135285914 136296092 0,223449 A T 22 chr9 135535698 135535698 136545876 0,404554 A G 23 chr9 135643869 rsl50630093 135643869 136654047 0,204194 T C 24 chr9 135648183 rsl39048080 135648183 136658361 0,265139 A G 25 chr9 135651645 rs41302911 135651645 136661823 0,204629 A G 26 chr9 135691191 135691191 136701369 0,356692 T C 27 chr9 135691447 135691447 136701625 0,450412 T C 28 chr9 135741616 135741616 136751794 0,45677 T C 29 chr9 135764472 135764472 136774650 0,495686 T C 30 chr9 135805303 135805303 136815481 0,22868 T C 31 chr9 135959983 135959983 136970161 0,243951 T C 32

Association analysis

For single marker association to a disease, the Fisher exact test can be used to calculate two- sided p-values for each individual allele. Correcting for relatedness among patients can be done by extending a variance adjustment procedure previously described (Risch, N. & Teng, J.

Genome Res., 8 : 1273- 1288 ( 1998)) for sibships so that it can be applied to general familial relationships. The method of genomic controls (Devlin, B. & Roeder, K. Biometrics 55 :997 ( 1999)) can also be used to adjust for the relatedness of the individuals and possible

stratification.

For both single-marker and haplotype analyses, relative risk (RR) and the population attributable risk (PAR) can be calculated assuming a multiplicative model (haplotype relative risk model) (Terwilliger, J. D. & Ott, J., Hum. Hered. 42 :337-46 (1992) and Falk, C.T. & Rubinstein, P, Ann. Hum. Genet. 51 (Pt 3) : 227-33 ( 1987)), i.e., that the risks of the two alleles/haplotypes a person carries multiply. For example, if RR is the risk of A relative to a, then the risk of a person homozygote AA will be RR times that of a heterozygote Aa and RR² times that of a homozygote aa . The multiplicative model has a nice property that simplifies analysis and computations— haplotypes are independent, i.e., in Hardy-Weinberg equilibrium, within the affected population as well as within the control population. As a consequence, haplotype counts of the affecteds and controls each have multinomial distributions, but with different haplotype frequencies under the alternative hypothesis. Specifically, for two haplotypes, /?, and h_j, risk(A7 )/risk(A7₇-) =

if p /ifj/Pj), where f and p denote, respectively, frequencies in the affected population and in the control population. While there is some power loss if the true model is not multiplicative, the loss tends to be mild except for extreme cases. Most importantly, p-values are always valid since they are computed with respect to null hypothesis.

An association signal detected in one association study may be replicated in a second cohort, for example a cohort from a different population (e.g., different region of same country, or a different country) of the same or different ethnicity. The advantage of replication studies is that the number of tests performed in the replication study is usually quite small, and hence the less stringent the statistical measure that needs to be applied. For example, for a genome-wide search for susceptibility variants for a particular disease or trait using 300,000 SNPs, a correction for the 300,000 tests performed (one for each SNP) can be performed. Since many SNPs on the arrays typically used are correlated (i.e., in LD), they are not independent. Thus, the correction is conservative. Nevertheless, applying this correction factor requires an observed P-value of less than 0.05/300,000 = 1.7 x 10^"7 for the signal to be considered significant applying this conservative test on results from a single study cohort. Obviously, signals found in a genome- wide association study with P-values less than this conservative threshold (i.e., more significant) are a measure of a true genetic effect, and replication in additional cohorts is not necessary from a statistical point of view. Importantly, however, signals with P-values that are greater than this threshold may also be due to a true genetic effect. The sample size in the first study may not have been sufficiently large to provide an observed P-value that meets the conservative threshold for genome-wide significance, or the first study may not have reached genome-wide significance due to inherent fluctuations due to sampling. Since the correction factor depends on the number of statistical tests performed, if one signal (one SNP) from an initial study is replicated in a second case-control cohort, the appropriate statistical test for significance is that for a single statistical test, i.e., P-value less than 0.05. Replication studies in one or even several additional case-control cohorts have the added advantage of providing assessment of the association signal in additional populations, thus simultaneously confirming the initial finding and providing an assessment of the overall significance of the genetic variant(s) being tested in human populations in general.

The results from several case-control cohorts can also be combined to provide an overall assessment of the underlying effect. The methodology commonly used to combine results from multiple genetic association studies is the Mantel-Haenszel model (Mantel and Haenszel, J Natl Cancer Inst 22:719-48 (1959)). The model is designed to deal with the situation where association results from different populations, with each possibly having a different population frequency of the genetic variant, are combined. The model combines the results assuming that the effect of the variant on the risk of the disease, a measured by the OR or RR, is the same in all populations, while the frequency of the variant may differ between the populations.

Combining the results from several populations has the added advantage that the overall power to detect a real underlying association signal is increased, due to the increased statistical power provided by the combined cohorts. Furthermore, any deficiencies in individual studies, for example due to unequal matching of cases and controls or population stratification will tend to balance out when results from multiple cohorts are combined, again providing a better estimate of the true underlying genetic effect.

Risk assessment and Diagnostics

Within any given population, there is an absolute risk of developing a disease or trait, defined as the chance of a person developing the specific disease or trait over a specified time-period. For example, a woman's lifetime absolute risk of breast cancer is one in nine. That is to say, one woman in every nine will develop breast cancer at some point in their lives. Risk is typically measured by looking at very large numbers of people, rather than at a particular individual. Risk is often presented in terms of Absolute Risk (AR) and Relative Risk (RR). Relative Risk is used to compare risks associating with two variants or the risks of two different groups of people. For example, it can be used to compare a group of people with a certain genotype with another group having a different genotype. For a disease, a relative risk of 2 means that one group has twice the chance of developing a disease as the other group. The risk presented is usually the relative risk for a person, or a specific genotype of a person, compared to the population with matched gender and ethnicity. Risks of two individuals of the same gender and ethnicity could be compared in a simple manner. For example, if, compared to the population, the first individual has relative risk 1.5 and the second has relative risk 0.5, then the risk of the first individual compared to the second individual is 1.5/0.5 = 3.

Risk Calculations

The creation of a model to calculate the overall genetic risk involves two steps: i) conversion of odds-ratios for a single genetic variant into relative risk and ii) combination of risk from multiple variants in different genetic loci into a single relative risk value.

Deriving risk from odds-ratios

Most gene discovery studies for complex diseases that have been published to date in authoritative journals have employed a case-control design because of their retrospective setup. These studies sample and genotype a selected set of cases (people who have the specified disease condition) and control individuals. The interest is in genetic variants (alleles) which frequency in cases and controls differ significantly.

The results are typically reported in odds ratios, that is the ratio between the fraction

(probability) with the risk variant (carriers) versus the non-risk variant (non-carriers) in the groups of affected versus the controls, i.e. expressed in terms of probabilities conditional on the affection status:

OR = (Pr(c|A)/Pr(nc|A)) / (Pr(c|C)/Pr(nc|C))

Sometimes it is however the absolute risk for the disease that we are interested in, i.e. the fraction of those individuals carrying the risk variant who get the disease or in other words the probability of getting the disease. This number cannot be directly measured in case-control studies, in part, because the ratio of cases versus controls is typically not the same as that in the general population. However, under certain assumption, we can estimate the risk from the odds ratio.

It is well known that under the rare disease assumption, the relative risk of a disease can be approximated by the odds ratio. This assumption may however not hold for many common diseases. Still, it turns out that the risk of one genotype variant relative to another can be estimated from the odds ratio expressed above. The calculation is particularly simple under the assumption of random population controls where the controls are random samples from the same population as the cases, including affected people rather than being strictly unaffected individuals. To increase sample size and power, many of the large genome-wide association and replication studies use controls that were neither age-matched with the cases, nor were they carefully scrutinized to ensure that they did not have the disease at the time of the study.

Hence, while not exactly, they often approximate a random sample from the general population. It is noted that this assumption is rarely expected to be satisfied exactly, but the risk estimates are usually robust to moderate deviations from this assumption.

Calculations show that for the dominant and the recessive models, where we have a risk variant carrier, "c", and a non-carrier, "nc", the odds ratio of individuals is the same as the risk ratio between these variants:

OR = Pr(A|c)/Pr(A| nc) = r

And likewise for the multiplicative model, where the risk is the product of the risk associated with the two allele copies, the allelic odds ratio equals the risk factor:

OR = Pr(A|aa)/Pr(A|ab) = Pr(A|ab)/Pr(A| bb) = r

Here "a" denotes the risk allele and "b" the non-risk allele. The factor "r" is therefore the relative risk between the allele types.

For many of the studies published in the last few years, reporting common variants associated with complex diseases, the multiplicative model has been found to summarize the effect adequately and most often provide a fit to the data superior to alternative models such as the dominant and recessive models.

Determining risk

In the present context, an individual who is at an increased susceptibility (i.e., increased risk) for Thyroid Cancer is an individual who is carrying at least one at-risk allele at particular genetic markers. In certain embodiments, the genetic markers are within the human RALGDS gene (e.g., rsl l3532379). Alternatively, an individual who is at an increased susceptibility for Thyroid Cancer is an individual who is carrying at least one at-risk allele in a marker that is correlated with such genetic markers (e.g., rsl l3532379). In one embodiment, significance associated with a marker is measured by a relative risk (RR). In another embodiment, significance associated with a marker or haplotye is measured by an odds ratio (OR). In a further embodiment, the significance is measured by a percentage. In one embodiment, a significant increased risk is measured as a risk (relative risk and/or odds ratio) of at least 1.5, including but not limited to: at least 2.0, at least 2.5, at least 3.0, at least 3.5, at least 4.0, at least 4.5, at least 5.0, at least 5.5, at least 6.0, at least 6.5, and at least 7.0. In a particular embodiment, a risk (relative risk and/or odds ratio) of at least 2.0 is significant. In another particular embodiment, a risk of at least 3.0 is significant.

An at-risk polymorphic marker as described herein is one where at least one allele of at least one marker is more frequently present in an individual diagnosed with, or at risk for, Thyroid Cancer (affected), compared to the frequency of its presence in a comparison group (control), such that the presence of the marker allele is indicative of increased susceptibility to Thyroid Cancer. The control group may in one embodiment be a population sample, i.e. a random sample from the general population. In another embodiment, the control group is represented by a group of individuals who are disease-free, i.e. individuals who have not been diagnosed with Thyroid

Cancer.

The person skilled in the art will appreciate that for markers with two alleles present in the population being studied (such as SNPs), and wherein one allele is found in increased frequency in a group of individuals with a trait or disease in the population, compared with controls, the other allele of the marker will be found in decreased frequency in the group of individuals with the trait or disease, compared with controls. In such a case, one allele of the marker (the one found in increased frequency in individuals with the trait or disease) will be the at-risk allele, while the other allele will be a protective allele.

Database

Determining susceptibility can alternatively or additionally comprise comparing nucleic acid sequence data and/or genotype data to a database containing correlation data between

polymorphic markers and susceptibility to Thyroid Cancer. The database can be part of a computer-readable medium described herein.

In a specific aspect of the invention, the database comprises at least one measure of

susceptibility to the condition for the polymorphic markers. For example, the database may comprise risk values associated with particular genotypes at such markers. The database may also comprise risk values associated with particular genotype combinations for multiple such markers.

In another specific aspect of the invention, the database comprises a look-up table containing at least one measure of susceptibility to the condition for the polymorphic markers.

Further steps

The methods disclosed herein can comprise additional steps which may occur before, after, or simultaneously with one of the aforementioned steps of the method of the invention. In a specific embodiment of the invention, the method of determining a susceptibility to Thyroid Cancer further comprises reporting the susceptibility to at least one entity selected from the group consisting of the individual, a guardian of the individual, a genetic service provider, a physician, a medical organization, and a medical insurer. The reporting may be accomplished by any of several means. For example, the reporting can comprise sending a written report on physical media or electronically or providing an oral report to at least one entity of the group, which written or oral report comprises the susceptibility. Alternatively, the reporting can comprise providing the at least one entity of the group with a login and password, which provides access to a report comprising the susceptibility posted on a password -protected computer system.

Study population

In a general sense, the methods and kits described herein can be utilized from samples

containing nucleic acid material (DNA or RNA) from any source and from any individual, or from genotype or sequence data derived from such samples. In preferred embodiments, the

individual is a human individual. The individual can be an adult, child, or fetus. The nucleic acid source may be any sample comprising nucleic acid material, including biological samples, or a sample comprising nucleic acid material derived therefrom. The present invention also provides for assessing markers in individuals who are members of a target population. Such a target population is in one embodiment a population or group of individuals at risk of developing Thyroid Cancer, based on other genetic factors, biomarkers, biophysical parameters, history of Thyroid Cancer, family history of Thyroid Cancer or a related disease. In certain embodiments, a target population is a population with abnormal levels (high or low) of TSH, T4 or T3.

The Icelandic population is a Caucasian population of Northern European ancestry. A large number of studies reporting results of genetic linkage and association in the Icelandic population have been published in the last few years. Many of those studies show replication of variants, originally identified in the Icelandic population as being associating with a particular disease, in other populations (Sulem, P., et al. Nat Genet May 17 2009 (Epub ahead of print); Rafnar, T., et al. Nat Genet 41 :221-7 (2009); Greta rsdottir, S., et al. Ann Neurol 64:402-9 (2008); Stacey, S.N., et al. Nat Genet 40: 1313-18 (2008); Gudbjartsson, D.F., et al. Nat Genet 40:886-91 (2008); Styrka rsdottir, U., et al. N Engl J Med 358:2355-65 (2008); Thorgeirsson, T., et al.

Nature 452:638-42 (2008); Gudmundsson, J., et al. Nat Genet. 40:281-3 (2008); Stacey, S.N., et al., Nat Genet. 39:865-69 (2007); Helgadottir, A., et al., Science 316: 1491-93 (2007);

Steinthorsdottir, V., et al., Nat Genet. 39:770-75 (2007); Gudmundsson, J., et al., Nat Genet. 39:631-37 (2007); Frayling, TM, Nature Reviews Genet 8:657-662 (2007); Amundadottir, L.T., et al., Nat Genet. 38:652-58 (2006); Grant, S.F., et al., Nat Genet. 38:320-23 (2006)). Thus, genetic findings in the Icelandic population have in general been replicated in other populations, including populations from Africa and Asia.

It is thus believed that the markers described herein to be associated with risk of Thyroid Cancer will show similar association in other human populations. Particular embodiments comprising individual human populations are thus also contemplated and within the scope of the invention. Such embodiments relate to human subjects that are from one or more human population including, but not limited to, Caucasian populations, European populations, American

populations, Eurasian populations, and Asian populations.

The racial contribution in individual subjects may also be determined by genetic analysis using methods known to the skilled person. Genetic analysis of ancestry may for example be carried out using unlinked microsatellite markers such as those set out in Smith et a/. (Am J Hum Genet 74, 1001-13 (2004)).

In certain embodiments, the invention relates to markers identified in specific populations, as described in the above. The person skilled in the art will appreciate that measures of linkage disequilibrium (LD) may give different results when applied to different populations. This is due to different population history of different human populations as well as differential selective pressures that may have led to differences in LD in specific genomic regions. It is also well known to the person skilled in the art that certain markers, e.g. SNP markers, have different population frequency in different populations, or are polymorphic in one population but not in another. The person skilled in the art will however apply the methods available and as taught herein to practice the present invention in any given human population. This may include assessment of polymorphic markers in the LD region of the present invention, so as to identify those markers that give strongest association within the specific population. Thus, the at-risk variants of the present invention may reside on different haplotype background and in different frequencies in various human populations. However, utilizing methods known in the art and the markers of the present invention, the invention can be practiced in any given human population.

Screening Methods

The invention also provides a method of screening candidate markers for assessing susceptibility to Thyroid Cancer. The invention also provides a method of identification of a marker for use in assessing susceptibility to Thyroid Cancer. The method may comprise analyzing the frequency of at least one allele of a polymorphic marker in a population of human individuals diagnosed with Thyroid Cancer, wherein a significant difference in frequency of the at least one allele in the population of human individuals diagnosed with Thyroid Cancer as compared to the frequency of the at least one allele in a control population of human individuals is indicative of the allele as a marker of the Thyroid Cancer. In certain embodiments, the candidate marker is a marker in the human RALGDS gene. In certain embodiments, the candidate marker is in linkage disequilibrium with marker rsl l3532379.

In one embodiment, the method comprises (i) identifying at least one polymorphic marker in the human RALGDS gene; (ii) obtaining sequence information about the at least one polymorphic marker in a group of individuals diagnosed with Thyroid Cancer; and (iii) obtaining sequence information about the at least one polymorphic marker in a group of control individuals; wherein determination of a significant difference in frequency of at least one allele in the at least one polymorphism in individuals diagnosed with Thyroid Cancer as compared with the frequency of the at least one allele in the control group is indicative of the at least one polymorphism being useful for assessing susceptibility to Thyroid Cancer.

In one embodiment, an increase in frequency of the at least one allele in the at least one polymorphism in individuals diagnosed with Thyroid Cancer, as compared with the frequency of the at least one allele in the control group, is indicative of the at least one polymorphism being useful for assessing increased susceptibility to Thyroid Cancer. In another embodiment, a decrease in frequency of the at least one allele in the at least one polymorphism in individuals diagnosed with Thyroid Cancer, as compared with the frequency of the at least one allele in the control group, is indicative of the at least one polymorphism being useful for assessing decreased susceptibility to, or protection against, Thyroid Cancer.

Thyroid stimulating hormone

Thyroid-stimulating hormone (also known as TSH or thyrotropin) is a peptide hormone synthesized and secreted by thyrotrope cells in the anterior pituitary gland which regulates the endocrine function of the thyroid gland. TSH stimulates the thyroid gland to secrete the hormones thyroxine (T₄) and triiodothyronine (T₃). TSH production is controlled by a

Thyrotropin Releasing Hormone, (TRH), which is manufactured in the hypothalamus and transported to the anterior pituitary gland via the superior hypophyseal artery, where it increases TSH production and release. Somatostatin is also produced by the hypothalamus, and has an opposite effect on the pituitary production of TSH, decreasing or inhibiting its release.

The level of thyroid hormones (T₃ and T₄) in the blood have an effect on the pituitary release of TSH; when the levels of T₃ and T₄ are low, the production of TSH is increased, and conversely, when levels of T₃ and T₄ are high, then TSH production is decreased. This effect creates a regulatory negative feedback loop.

Thyroxine, or 3,5,3',5'-tetraiodothyronine (often abbreviated as T₄), is the major hormone secreted by the follicular cells of the thyroid gland. T₄ is transported in blood, with 99.95% of the secreted T₄ being protein bound, principally to thyroxine-binding globulin (TBG), and, to a lesser extent, to transthyretin and serum albumin. T₄ is involved in controlling the rate of metabolic processes in the body and influencing physical development. Administration of thyroxine has been shown to significantly increase the concentration of nerve growth factor in the brains of adult mice.

In the hypothalamus, T₄ is converted to Triiodothyronine, also known as T₃. TSH is inhibited mainly by T₃. The thyroid gland releases greater amounts of T₄ than T₃, so plasma

concentrations of T₄ are 40-fold higher than those of T₃. Most of the circulating T₃ is formed peripherally by deiodination of T₄ (85%), a process that involves the removal of iodine from carbon 5 on the outer ring of T₄. Thus, T₄ acts as prohormone for T₃. Utility of Genetic Testing

As discussed in the above, the primary known risk factor for thyroid cancer is radiation exposure.. Thyroid cancer incidence within the US has been rising for several decades (Davies, L. and Welch, H. G., Jama, 295, 2164 (2006)), which may be attributable to increased detection of sub-clinical cancers, as opposed to an increase in the true occurrence of thyroid cancer (Davies, L. and Welch, H. G., Jama, 295, 2164 (2006)). The introduction of ultrasonography and fine- needle aspiration biopsy in the 1980s improved the detection of small nodules and made cytological assessment of a nodule more routine (Rojeski, M. T. and Gharib, H., N Engl J Med, 313, 428 (1985), Ross, D. S., J Clin Endocrinol Metab, 91, 4253 (2006)). This increased diagnostic scrutiny may allow early detection of potentially lethal thyroid cancers. However, several studies report thyroid cancers as a common autopsy finding (up to 35%) in persons without a diagnosis of thyroid cancer ( Bondeson, L. and Ljungberg, O., Cancer, 47, 319 (1981), Harach, H. R., et al., Cancer, 56, 531 (1985), Solares, C. A., et a/., Am J Otolaryngol, 26, 87 (2005) and Sobrinho-Simoes, M. A., Sambade, M. C, and Goncalves, V., Cancer, 43, 1702 (1979)). This suggests that many people live with sub-clinical forms of thyroid cancer which are of little or no threat to their health.

Physicians use several tests to confirm the suspicion of thyroid cancer, to identify the size and location of the lump and to determine whether the lump is non-cancerous (benign) or cancerous (malignant). Blood tests such as the thyroid stimulating hormone (TSH) test check thyroid function. TSH levels are tested in the blood of patients suspected of suffering from excess (hyperthyroidism), or deficiency (hypothyroidism) of thyroid hormone. Generally, a normal range for TSH for adults is between 0.2 and 10 uIU/mL (equivalent to mlU/L). The optimal TSH level for patients on treatment ranges between 0.3 to 3.0 mlU/L. The interpretation of TSH measurements depends also on what the blood levels of thyroid hormones (T₃ and T₄) are. The National Health Service in the UK considers a "normal" range to be more like 0.1 to 5.0 uIU/mL.

TSH levels for children normally start out much higher. In 2002, the National Academy of Clinical Biochemistry (NACB) in the United States recommended age-related reference limits starting from about 1.3-19 uIU/mL for normal term infants at birth, dropping to 0.6-10 uIU/mL at 10 weeks old, 0.4-7.0 uIU/mL at 14 months and gradually dropping during childhood and puberty to adult levels, 0.4-4.0 uIU/mL. The NACB also stated that it expected the normal (95%) range for adults to be reduced to 0.4-2.5 uIU/mL, because research had shown that adults with an initially measured TSH level of over 2.0 uIU/mL had an increased odds ratio of developing hypothyroidism over the [following] 20 years, especially if thyroid antibodies were elevated.

In general, both TSH and T₃ and T₄ should be measured to ascertain where a specific thyroid dysfunction is caused by primary pituitary or by a primary thyroid disease. If both are up (or down) then the problem is probably in the pituitary. If the one component (TSH) is up, and the other (T₃ and T₄) is down, then the disease is probably in the thyroid itself. The same holds for a low TSH, high T3 and T4 finding.

The knowledge of underlying genetic risk factors for thyroid cancer can be utilized in the application of screening programs for thyroid cancer. Thus, carriers of at-risk variants for thyroid cancer may benefit from more frequent screening than do non-carriers. Homozygous carriers of at-risk variants are particularly at risk for developing thyroid cancer.

It may be beneficial to determine TSH, T3 and/or T4 levels in the context of a particular genetic profile, e.g. the presence of particular at-risk alleles for thyroid cancer as described herein (e.g., at-risk alleles in the human RALGDS gene; e.g., rsl l3532379 allele T). Since TSH, T3 and T4 are measures of thyroid function, a diagnostic and preventive screening program will benefit from analysis that includes such clinical measurements. For example, an abnormal (increased or decreased) level of TSH together with determination of the presence of an at-risk genetic variant for thyroid cancer (e.g., rsl 13532379) is indicative that an individual is at risk of developing thyroid cancer. In one embodiment, determination of a decreased level of TSH in an individual in the context of the presence of rsl 13532379 allele T is indicative of an increased risk of thyroid cancer for the individual.

Also, carriers may benefit from more extensive screening, including ultrasonography and /or fine needle biopsy. The goal of screening programs is to detect cancer at an early stage. Knowledge of genetic status of individuals with respect to known risk variants can aid in the selection of applicable screening programs. In certain embodiments, it may be useful to use the at-risk variants for thyroid cancer described herein together with one or more diagnostic tool selected from Radioactive Iodine (RAI) Scan, Ultrasound examination, CT scan (CAT scan), Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET) scan, Fine needle aspiration biopsy and surgical biopsy.

The invention provides in one diagnostic aspect a method for identifying a subject who is a candidate for further diagnostic evaluation for thyroid cancer, comprising the steps of (a) determining, in the genome of a human subject, the allelic identity of at least one polymorphic marker, wherein different alleles of the at least one marker are associated with different susceptibilities to thyroid cancer, and wherein the at least one marker is a marker in the human RALGDS gene; and (b) identifying the subject as a subject who is a candidate for further diagnostic evaluation for thyroid cancer based on the allelic identity at the at least one polymorphic marker. In one embodiment, the marker is rsl l3532379, wherein a determination of the presence of the T allele of rsl 13532379 is indicative that the individual is a candidate for further diagnostic evaluation for thyroid cancer. Thus, the identification of individuals who are at increased risk of developing thyroid cancer may be used to select those individuals for follow-up clinical evaluation, as described in the above.

Prognostic methods

In addition to the utilities described above, the polymorphic markers of the invention are useful in determining prognosis of a human individual experiencing symptoms associated with, or an individual diagnosed with, thyroid cancer. Accordingly, the invention provides a method of predicting prognosis of an individual experiencing symptoms associated with, or an individual diagnosed with, thyroid cancer. The method comprises analyzing data about at least one allele of human RALGDS gene (SEQ ID NO: 57) in a human subject, wherein different alleles of the human RALGDS gene are associated with different susceptibilities to thyroid cancer in humans, and predicting prognosis of the individual from the sequence data. In certain embodiments, the at least one allele is the T allele of rsl l3532379.

The prognosis can be any type of prognosis relating to the progression of thyroid cancer, and/or relating to the chance of recovering from thyroid cancer. The prognosis can, for instance, relate to the severity of the cancer, when the cancer may take place (e.g., the likelihood of

recurrence), or how the cancer will respond to therapeutic treatment.

With regard to the prognostic methods described herein, the sequence data obtained to establish a prognostic prediction is suitably nucleic acid sequence data. For example, in one embodiment, determination of the presence of an at-risk allele of thyroid cancer (e.g., rsl 13532379 allele T) is useful for prognostic applications. Suitable methods of detecting particular at-risk alleles are known in the art, some of which are described herein.

Kits

Kits useful in the methods of the invention comprise components useful in any of the methods described herein, including for example, primers for nucleic acid amplification, hybridization probes, restriction enzymes (e.g., for RFLP analysis), allele-specific oligonucleotides, antibodies, means for amplification of nucleic acids, means for analyzing the nucleic acid sequence of nucleic acids, means for analyzing the amino acid sequence of a polynucleotides, etc. The kits can for example include necessary buffers, nucleic acid primers for amplifying nucleic acids (e.g., a nucleic acid segment comprising one or more of the polymorphic markers as described herein), and reagents for allele-specific detection of the fragments amplified using such primers and necessary enzymes (e.g. , DNA polymerase). Additionally, kits can provide reagents for assays to be used in combination with the methods of the present invention, e.g., reagents for use with other diagnostic assays for thyroid cancer.

In one embodiment, the invention pertains to a kit for assaying a sample from a subject to detect a susceptibility to thyroid cancer in the subject, wherein the kit comprises reagents necessary for selectively detecting at least one allele of the human RALGDS gene (SEQ ID NOXX) in the subject. In one embodiment, the at least one allele is the T allele of rsl l3532379. In another embodiment, the at least one allele is the G allele of rsl39082000. In a particular embodiment, the reagents comprise at least one contiguous oligonucleotide that hybridizes to a fragment of the genome of the individual comprising at least one polymorphism as described herein. In another embodiment, the reagents comprise at least one pair of oligonucleotides that hybridize to opposite strands of a genomic segment obtained from a subject, wherein each oligonucleotide primer pair is designed to selectively amplify a fragment of the genome of the individual that includes at least one allele associated with thyroid cancer risk. In one such embodiment, the allele is the T allele or rsl l3532379, or polymorphic marker allele in linkage disequilibrium therewith. In yet another embodiment the fragment is at least 20 base pairs in size. Such oligonucleotides or nucleic acids (e.g. , oligonucleotide primers) can be designed using portions of the nucleic acid sequence flanking the polymorphism. In another embodiment, the kit comprises one or more labeled nucleic acids capable of allele-specific detection of one or more specific polymorphic markers or haplotypes, and reagents for detection of the label.

Suitable labels include, e.g., a radioisotope, a fluorescent label, an enzyme label, an enzyme co- factor label, a magnetic label, a spin label, an epitope label.

In one embodiment, the DNA template is amplified before detection by PCR. The DNA template may also be amplified by means of Whole Genome Amplification (WGA) methods, prior to assessment for the presence of specific polymorphic markers as described herein. Standard methods well known to the skilled person for performing WGA may be utilized, and are within scope of the invention. In one such embodiment, reagents for performing WGA are included in the reagent kit.

In certain embodiments, determination of the presence of a particular marker allele (e.g. allele T of rsl 13532379) is indicative of increased susceptibility of thyroid cancer. In another embodiment, determination of the presence of a particular marker allele is indicative of prognosis of thyroid cancer. In another embodiment, the presence of a marker allele is indicative of response to a therapeutic agent for thyroid cancer. In yet another embodiment, the presence of a marker allele is indicative of progress of treatment of thyroid cancer.

In certain embodiments, the kit comprises reagents for detecting no more than 100 alleles in the genome of the individual. In certain other embodiments, the kit comprises reagents for detecting no more than 20 alleles in the genome of the individual. In a further aspect of the present invention, a pharmaceutical pack (kit) is provided, the pack comprising a therapeutic agent and a set of instructions for administration of the therapeutic agent to humans diagnostically tested for an at-risk variant for thyroid cancer. The therapeutic agent can be a small molecule drug, an antibody, a peptide, an antisense or RNAi molecule, or other therapeutic molecules. In one embodiment, an individual identified as a carrier of at least one variant of the present invention is instructed to take a prescribed dose of the therapeutic agent. In one such embodiment, an individual identified as a homozygous carrier of at least one variant of the present invention (e.g., an at-risk variant) is instructed to take a prescribed dose of the therapeutic agent. In another embodiment, an individual identified as a non-carrier of at least one variant of the present invention (e.g., an at-risk variant) is instructed to take a prescribed dose of the therapeutic agent.

In certain embodiments, the kit further comprises a set of instructions for using the reagents comprising the kit. In certain embodiments, the kit further comprises a collection of data comprising correlation data between the at least one at-risk variant and susceptibility to thyroid cancer.

Nucleic acids and polypeptides

The nucleic acids and polypeptides described herein can be used in methods and kits of the present invention. An "isolated" nucleic acid molecule, as used herein, is one that is separated from nucleic acids that normally flank the gene or nucleotide sequence (as in genomic sequences) and/or has been completely or partially purified from other transcribed sequences (e.g., as in an RNA library). For example, an isolated nucleic acid of the invention can be substantially isolated with respect to the complex cellular milieu in which it naturally occurs, or culture medium when produced by recombinant techniques, or chemical precursors or other chemicals when chemically synthesized. In some instances, the isolated material will form part of a composition (for example, a crude extract containing other substances), buffer system or reagent mix. In other circumstances, the material can be purified to essential homogeneity, for example as determined by polyacrylamide gel electrophoresis (PAGE) or column chromatography (e.g., HPLC). An isolated nucleic acid molecule of the invention can comprise at least about 50%, at least about 80% or at least about 90% (on a molar basis) of all macromolecular species present. With regard to genomic DNA, the term "isolated" also can refer to nucleic acid molecules that are separated from the chromosome with which the genomic DNA is naturally associated. For example, the isolated nucleic acid molecule can contain less than about 250 kb, 200 kb, 150 kb, 100 kb, 75 kb, 50 kb, 25 kb, 10 kb, 5 kb, 4 kb, 3 kb, 2 kb, 1 kb, 0.5 kb or 0.1 kb of the nucleotides that flank the nucleic acid molecule in the genomic DNA of the cell from which the nucleic acid molecule is derived.

The invention also pertains to nucleic acid molecules that hybridize under high stringency hybridization conditions, such as for selective hybridization, to a nucleotide sequence described herein (e.g., nucleic acid molecules that specifically hybridize to a nucleotide sequence containing a polymorphic site associated with a marker or haplotype described herein). Such nucleic acid molecules can be detected and/or isolated by allele- or sequence-specific hybridization (e.g., under high stringency conditions). Stringency conditions and methods for nucleic acid hybridizations are well known to the skilled person (see, e.g., Current Protocols in Molecular Biology, Ausubel, F. et al, John Wiley & Sons, (1998), and Kraus, M. and Aaronson, S., Methods Enzymol., 200: 546-556 (1991), the entire teachings of which are incorporated by reference herein.

The percent identity of two nucleotide or amino acid sequences can be determined by aligning the sequences for optimal comparison purposes (e.g., gaps can be introduced in the sequence of a first sequence). The nucleotides or amino acids at corresponding positions are then compared, and the percent identity between the two sequences is a function of the number of identical positions shared by the sequences (i.e., % identity = # of identical positions/total # of positions x 100). In certain embodiments, the length of a sequence aligned for comparison purposes is at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or at least 95%, of the length of the reference sequence. The actual comparison of the two sequences can be accomplished by well-known methods, for example, using a mathematical algorithm. A non-limiting example of such a mathematical algorithm is described in Karlin, S. and Altschul, S., Proc. Natl. Acad. Sci. USA, 90:5873-5877 (1993). Such an algorithm is incorporated into the NBLAST and XBLAST programs (version 2.0), as described in Altschul, S. et al., Nucleic Acids Res., 25:3389-3402 (1997). When utilizing BLAST and Gapped BLAST programs, the default parameters of the respective programs (e.g., NBLAST) can be used. See the website on the world wide web at ncbi.nlm.nih.gov. In one embodiment, parameters for sequence comparison can be set at score= 100, wordlength = 12, or can be varied (e.g., W=5 or W=20). Another example of an algorithm is BLAT (Kent, WJ. Genome Res. 12:656-64 (2002)).

Other examples include the algorithm of Myers and Miller, CABIOS (1989), ADVANCE and ADAM as described in Torellis, A. and Robotti, C, Comput. Appl. Biosci. 10:3-5 (1994); and FASTA described in Pearson, W. and Lipman, D., Proc. Natl. Acad. Sci. USA, 85:2444-48 (1988). In another embodiment, the percent identity between two amino acid sequences can be

accomplished using the GAP program in the GCG software package (Accelrys, Cambridge, UK).

The present invention also provides isolated nucleic acid molecules that contain a fragment or portion that hybridizes under highly stringent conditions to a nucleic acid that comprises, or consists of, the nucleotide sequence of the human RALGDS gene as set forth SEQ ID NO: 57, or a nucleotide sequence comprising, or consisting of, the complement of the nucleotide sequence of SEQ ID NO: 57. The nucleic acid fragments of the invention are suitably at least about 15, at least about 18, 20, 23 or 25 nucleotides, and can be up to 30, 40, 50, 100, 200, 300 or 400 nucleotides in length.

The nucleic acid fragments of the invention are used as probes or primers in assays such as those described herein. "Probes" or "primers" are oligonucleotides that hybridize in a base- specific manner to a complementary strand of a nucleic acid molecule. In addition to DNA and RNA, such probes and primers include polypeptide nucleic acids (PNA), as described in Nielsen, P. eta/., Science 254: 1497-1500 (1991). A probe or primer comprises a region of nucleotide sequence that hybridizes to at least about 15, typically about 20-25, and in certain embodiments about 40, 50 or 75, consecutive nucleotides of a nucleic acid molecule. In one embodiment, the probe or primer comprises at least one allele of at least one polymorphic marker or at least one haplotype described herein, or the complement thereof. In particular embodiments, a probe or primer can comprise 100 or fewer nucleotides; for example, in certain embodiments from 6 to 50 nucleotides, or, for example, from 12 to 30 nucleotides. In other embodiments, the probe or primer is at least 70% identical, at least 80% identical, at least 85% identical, at least 90% identical, or at least 95% identical, to the contiguous nucleotide sequence or to the complement of the contiguous nucleotide sequence. In another embodiment, the probe or primer is capable of selectively hybridizing to the contiguous nucleotide sequence or to the complement of the contiguous nucleotide sequence. Often, the probe or primer further comprises a label, e.g., a radioisotope, a fluorescent label, an enzyme label, an enzyme co-factor label, a magnetic label, a spin label, an epitope label.

Computer-implemented aspects

As understood by those of ordinary skill in the art, the methods and information described herein may be implemented, in all or in part, as computer executable instructions on known computer readable media. For example, the methods described herein may be implemented in hardware. Alternatively, the method may be implemented in software stored in, for example, one or more memories or other computer readable medium and implemented on one or more processors. As is known, the processors may be associated with one or more controllers, calculation units and/or other units of a computer system, or implanted in firmware as desired. If implemented in software, the routines may be stored in any computer readable memory such as in RAM, ROM, flash memory, a magnetic disk, a laser disk, or other storage medium, as is also known.

Likewise, this software may be delivered to a computing device via any known delivery method including, for example, over a communication channel such as a telephone line, the Internet, a wireless connection, etc., or via a transportable medium, such as a computer readable disk, flash drive, etc.

More generally, and as understood by those of ordinary skill in the art, the various steps described above may be implemented as various blocks, operations, tools, modules and techniques which, in turn, may be implemented in hardware, firmware, software, or any combination of hardware, firmware, and/or software. When implemented in hardware, some or all of the blocks, operations, techniques, etc. may be implemented in, for example, a custom integrated circuit (IC), an application specific integrated circuit (ASIC), a field programmable logic array (FPGA), a programmable logic array (PLA), etc.

When implemented in software, the software may be stored in any known computer readable medium such as on a magnetic disk, an optical disk, or other storage medium, in a RAM or ROM or flash memory of a computer, processor, hard disk drive, optical disk drive, tape drive, etc. Likewise, the software may be delivered to a user or a computing system via any known delivery method including, for example, on a computer readable disk or other transportable computer storage mechanism. Thus, another aspect of the invention is a system that is capable of carrying out a part or all of a method of the invention, or carrying out a variation of a method of the invention as described herein in greater detail. Exemplary systems include, as one or more components, computing systems, environments, and/or configurations that may be suitable for use with the methods and include, but are not limited to, personal computers, server computers, hand-held or laptop devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like. In some variations, a system of the invention includes one or more machines used for analysis of biological material (e.g., genetic material), as described herein. In some variations, this analysis of the biological material involves a chemical analysis and/or a nucleic acid amplification.

With reference to Fig. 1, an exemplary system of the invention, which may be used to implement one or more steps of methods of the invention, includes a computing device in the form of a computer 110. Components shown in dashed outline are not technically part of the computer 110, but are used to illustrate the exemplary embodiment of Fig. 1. Components of computer 110 may include, but are not limited to, a processor 120, a system memory 130, a

memory/graphics interface 121, also known as a Northbridge chip, and an I/O interface 122, also known as a Southbridge chip. The system memory 130 and a graphics processor 190 may be coupled to the memory/graphics interface 121. A monitor 191 or other graphic output device may be coupled to the graphics processor 190.

A series of system busses may couple various system components including a high speed system bus 123 between the processor 120, the memory/graphics interface 121 and the I/O interface 122, a front-side bus 124 between the memory/graphics interface 121 and the system memory 130, and an advanced graphics processing (AGP) bus 125 between the memory/graphics interface 121 and the graphics processor 190. The system bus 123 may be any of several types of bus structures including, by way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus and Enhanced ISA (EISA) bus. As system architectures evolve, other bus architectures and chip sets may be used but often generally follow this pattern. For example, companies such as Intel and AMD support the Intel Hub Architecture (IHA) and the Hypertransport™ architecture, respectively.

The computer 110 typically includes a variety of computer-readable media. Computer-readable media can be any available media that can be accessed by computer 110 and includes both volatile and nonvolatile media, removable and non-removable media. By way of example, and not limitation, computer readable media may comprise computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other physical medium which can be used to store the desired information and which can accessed by computer 110. The system memory 130 includes computer storage media in the form of volatile and/or nonvolatile memory such as read only memory (ROM) 131 and random access memory (RAM) 132. The system ROM 131 may contain permanent system data 143, such as identifying and manufacturing information. In some embodiments, a basic input/output system (BIOS) may also be stored in system ROM 131. RAM 132 typically contains data and/or program modules that are immediately accessible to and/or presently being operated on by processor 120. By way of example, and not limitation, Fig. 1 illustrates operating system 134, application programs 135, other program modules 136, and program data 137.

The I/O interface 122 may couple the system bus 123 with a number of other busses 126, 127 and 128 that couple a variety of internal and external devices to the computer 110. A serial peripheral interface (SPI) bus 126 may connect to a basic input/output system (BIOS) memory 133 containing the basic routines that help to transfer information between elements within computer 110, such as during start-up.

A super input/output chip 160 may be used to connect to a number of 'legacy' peripherals, such as floppy disk 152, keyboard/mouse 162, and printer 196, as examples. The super I/O chip 160 may be connected to the I/O interface 122 with a bus 127, such as a low pin count (LPC) bus, in some embodiments. Various embodiments of the super I/O chip 160 are widely available in the commercial marketplace.

In one embodiment, bus 128 may be a Peripheral Component Interconnect (PCI) bus, or a variation thereof, may be used to connect higher speed peripherals to the I/O interface 122. A PCI bus may also be known as a Mezzanine bus. Variations of the PCI bus include the Peripheral Component Interconnect-Express (PCI-E) and the Peripheral Component Interconnect - Extended (PCI-X) busses, the former having a serial interface and the latter being a backward compatible parallel interface. In other embodiments, bus 128 may be an advanced technology attachment (ATA) bus, in the form of a serial ATA bus (SATA) or parallel ATA (PATA).

The computer 110 may also include other removable/non-removable, volatile/nonvolatile computer storage media. By way of example only, Fig. 1 illustrates a hard disk drive 140 that reads from or writes to non-removable, nonvolatile magnetic media. The hard disk drive 140 may be a conventional hard disk drive.

Removable media, such as a universal serial bus (USB) memory 153, firewire (IEEE 1394), or CD/DVD drive 156 may be connected to the PCI bus 128 directly or through an interface 150. A storage media 154 may be coupled through interface 150. Other removable/non-removable, volatile/nonvolatile computer storage media that can be used in the exemplary operating environment include, but are not limited to, magnetic tape cassettes, flash memory cards, digital versatile disks, digital video tape, solid state RAM, solid state ROM, and the like.

The drives and their associated computer storage media discussed above and illustrated in Fig. 1, provide storage of computer readable instructions, data structures, program modules and other data for the computer 110. In Fig. 1, for example, hard disk drive 140 is illustrated as storing operating system 144, application programs 145, other program modules 146, and program data 147. Note that these components can either be the same as or different from operating system 134, application programs 135, other program modules 136, and program data 137. Operating system 144, application programs 145, other program modules 146, and program data 147 are given different numbers here to illustrate that, at a minimum, they are different copies. A user may enter commands and information into the computer 20 through input devices such as a mouse/keyboard 162 or other input device combination. Other input devices (not shown) may include a microphone, joystick, game pad, satellite dish, scanner, or the like. These and other input devices are often connected to the processor 120 through one of the I/O interface busses, such as the SPI 126, the LPC 127, or the PCI 128, but other busses may be used. In some embodiments, other devices may be coupled to parallel ports, infrared interfaces, game ports, and the like (not depicted), via the super I/O chip 160.

The computer 110 may operate in a networked environment using logical connections to one or more remote computers, such as a remote computer 180 via a network interface controller (NIC) 170, . The remote computer 180 may be a personal computer, a server, a router, a network PC, a peer device or other common network node, and typically includes many or all of the elements described above relative to the computer 110. The logical connection between the NIC 170 and the remote computer 180 depicted in Fig. 1 may include a local area network (LAN), a wide area network (WAN), or both, but may also include other networks. Such networking environments are commonplace in offices, enterprise-wide computer networks, intranets, and the Internet. The remote computer 180 may also represent a web server supporting interactive sessions with the computer 110, or in the specific case of location-based applications may be a location server or an application server.

In some embodiments, the network interface may use a modem (not depicted) when a broadband connection is not available or is not used. It will be appreciated that the network connection shown is exemplary and other means of establishing a communications link between the computers may be used.

In some variations, the invention is a system for identifying susceptibility to a cancer in a human subject. For example, in one variation, the system includes tools for performing at least one step, preferably two or more steps, and in some aspects all steps of a method of the invention, where the tools are operably linked to each other. Operable linkage describes a linkage through which components can function with each other to perform their purpose.

In some variations, a system of the invention is a system for identifying susceptibility to thyroid cancer in a human subject, and comprises:

(a) at least one processor;

(b) at least one computer-readable medium;

(c) a susceptibility database operatively coupled to a computer-readable medium of the system and containing population information correlating the presence or absence of one or more alleles of the human RALGDS gene and susceptibility to thyroid cancer in a population of humans; (d) a measurement tool that receives an input about the human subject and generates information from the input about the presence or absence of at least one mutant RALGDS allele indicative of a RALGDS defect in the human subject; and

(e) an analysis tool or routine that:

(i) is operatively coupled to the susceptibility database and the information generated by the measurement tool,

(ii) is stored on a computer-readable medium of the system,

(iii) is adapted to be executed on a processor of the system, to compare the information about the human subject with the population information in the susceptibility database and generate a conclusion with respect to susceptibility to thyroid cancer for the human subject.

Exemplary processors (processing units) include all variety of microprocessors and other processing units used in computing devices. Exemplary computer-readable media are described above. When two or more components of the system involve a processor or a computer- readable medium, the system generally can be created where a single processor and/or computer readable medium is dedicated to a single component of the system; or where two or more functions share a single processor and/or share a single computer readable medium, such that the system contains as few as one processor and/or one computer readable medium. In some variations, it is advantageous to use multiple processors or media, for example, where it is convenient to have components of the system at different locations. For instance, some components of a system may be located at a testing laboratory dedicated to laboratory or data analysis, whereas other components, including components (optional) for supplying input information or obtaining an output communication, may be located at a medical treatment or counseling facility (e.g., doctor's office, health clinic, HMO, pharmacist, geneticist, hospital) and/or at the home or business of the human subject (patient) for whom the testing service is performed.

Referring to Figure 2, an exemplary system includes a susceptibility database 208 that is operatively coupled to a computer-readable medium of the system and that contains population information correlating the presence or absence of one or more alleles of the human RALGDS gene and susceptibility to a cancer in a population of humans. For example, the one or more alleles of the RALGDS gene include mutant alleles that cause, or are indicative of, a RALGDS defect such as reduced or lost function, as described elsewhere herein.

In a simple variation, the susceptibility database contains 208 data relating to the frequency that a particular allele of RALGDS has been observed in a population of humans with thyroid cancer and a population of humans free of thyroid cancer. Such data provides an indication as to the relative risk or odds ratio of developing thyroid cancer for a human subject that is identified as having the allele in question. In another variation, the susceptibility database includes similar data with respect to two or more alleles of RALGDS, thereby providing a useful reference if the human subject has any of the two or more alleles. In still another variation, the susceptibility database includes additional quantitative personal, medical, or genetic information about the individuals in the database diagnosed with thyroid cancer or free of thyroid cancer. Such information includes, but is not limited to, information about parameters such as age, sex, ethnicity, race, medical history, weight, diabetes status, blood pressure, family history of the cancer, smoking history, and alcohol use in humans and impact of the at least one parameter on susceptibility to the cancer. The information also can include information about other genetic risk factors for the cancer besides RALGDS. These more robust susceptibility databases can be used by an analysis routine 210 to calculate a combined score with respect to susceptibility or risk for developing thyroid cancer.

In addition to the susceptibility database 208, the system further includes a measurement tool 206 programmed to receive an input 204 from or about the human subject and generate an output that contains information about the presence or absence of the at least one RALGDS allele of interest. (The input 204 is not part of the system per se but is illustrated in the schematic Figure 2.) Thus, the input 204 will contain a specimen or contain data from which the presence or absence of the at least one RALGDS allele can be directly read, or analytically determined. In a simple variation, the input contains annotated information about genotypes or allele counts for RALGDS in the genome of the human subject, in which case no further processing by the measurement tool 206 is required, except possibly transformation of the relevant information about the presence/absence of the RALGDS allele into a format compatible for use by the analysis routine 210 of the system.

In another variation, the input 204 from the human subject contains data that is unannotated or insufficiently annotated with respect to RALGDS, requiring analysis by the measurement tool 206. For example, the input can be genetic sequence of a chromosomal region or chromosome on which RALGDS resides, or whole genome sequence information, or unannotated information from a gene chip analysis of a variable loci in the human subject's genome. In such variations of the invention, the measurement tool 206 comprises a tool, preferably stored on a computer- readable medium of the system and adapted to be executed on a processor of the system, to receive a data input about a subject and determine information about the presence or absence of the at least one mutant RALGDS allele in a human subject from the data. For example, the measurement tool 206 contains instructions, preferably executable on a processor of the system, for analyzing the unannotated input data and determining the presence or absence of the RALGDS allele of interest in the human subject. Where the input data is genomic sequence information, and the measurement tool optionally comprises a sequence analysis tool stored on a computer readable medium of the system and executable by a processor of the system with instructions for determining the presence or absence of the at least one mutant RALGDS allele from the genomic sequence information.

In yet another variation, the input 204 from the human subject comprises a biological sample, such as a fluid (e.g., blood) or tissue sample that contains genetic material that can be analyzed to determine the presence or absence of the RALGDS allele of interest. In this variation, an exemplary measurement tool 206 includes laboratory equipment for processing and analyzing the sample to determine the presence or absence (or identity) of the RALGDS allele(s) in the human subject. For instance, in one variation, the measurement tool includes: an oligonucleotide microarray (e.g., "gene chip") containing a plurality of oligonucleotide probes attached to a solid support; a detector for measuring interaction between nucleic acid obtained from or amplified from the biological sample and one or more oligonucleotides on the

oligonucleotide microarray to generate detection data; and an analysis tool stored on a computer-readable medium of the system and adapted to be executed on a processor of the system, to determine the presence or absence of the at least one RALGDS allele of interest based on the detection data.

To provide another example, in some variations the measurement tool 206 includes: a nucleotide sequencer (e.g., an automated DNA sequencer) that is capable of determining nucleotide sequence information from nucleic acid obtained from or amplified from the biological sample; and an analysis tool stored on a computer-readable medium of the system and adapted to be executed on a processor of the system, to determine the presence or absence of the at least one mutant RALGDS allele based on the nucleotide sequence information.

In some variations, the measurement tool 206 further includes additional equipment and/or chemical reagents for processing the biological sample to purify and/or amplify nucleic acid of the human subject for further analysis using a sequencer, gene chip, or other analytical equipment.

The exemplary system further includes an analysis tool or routine 210 that: is operatively coupled to the susceptibility database 208 and operatively coupled to the measurement tool 206, is stored on a computer-readable medium of the system, is adapted to be executed on a processor of the system to compare the information about the human subject with the population information in the susceptibility database 208 and generate a conclusion with respect to susceptibility to thyroid cancer for the human subject. In simple terms, the analysis tool 210 looks at the RALGDS alleles identified by the measurement tool 206 for the human subject, and compares this information to the susceptibility database 208, to determine a susceptibility to thyroid cancer for the subject. The susceptibility can be based on the single parameter (the identity of one or more RALGDS alleles), or can involve a calculation based on other genetic and non-genetic data, as described above, that is collected and included as part of the input 204 from the human subject, and that also is stored in the susceptibility database 208 with respect to a population of other humans. Generally speaking, each parameter of interest is weighted to provide a conclusion with respect to susceptibility to thyroid cancer. Such a conclusion is expressed in the conclusion in any statistically useful form, for example, as an odds ratio, a relative risk, or a lifetime risk for subject developing thyroid cancer.

In some variations of the invention, the system as just described further includes a

communication tool 212. For example, the communication tool is operatively connected to the analysis routine 210 and comprises a routine stored on a computer-readable medium of the system and adapted to be executed on a processor of the system, to: generate a communication containing the conclusion; and to transmit the communication to the human subject 200 or the medical practitioner 202, and/or enable the subject or medical practitioner to access the communication. (The subject and medical practitioner are depicted in the schematic Fig. 2, but are not part of the system per se, though they may be considered users of the system. The communication tool 212 provides an interface for communicating to the subject, or to a medical practitioner for the subject (e.g., doctor, nurse, genetic counselor), the conclusion generated by the analysis tool 210 with respect to susceptibility to the cancer for the subject. Usually, if the communication is obtained by or delivered to the medical practitioner 202, the medical practitioner will share the communication with the human subject 200 and/or counsel the human subject about the medical significance of the communication. In some variations, the communication is provided in a tangible form, such as a printed report or report stored on a computer readable medium such as a flash drive or optical disk. In some variations, the communication is provided electronically with an output that is visible on a video display or audio output (e.g., speaker). In some variations, the communication is transmitted to the subject or the medical practitioner, e.g., electronically or through the mail. In some variations, the system is designed to permit the subject or medical practitioner to access the communication, e.g., by telephone or computer. For instance, the system may include software residing on a memory and executed by a processor of a computer used by the human subject or the medical practitioner, with which the subject or practitioner can access the communication, preferably securely, over the internet or other network connection. In some variations of the system, this computer will be located remotely from other components of the system, e.g., at a location of the human subject's or medical practitioner's choosing.

In some variations of the invention, the system as described (including embodiments with or without the communication tool) further includes components that add a treatment or prophylaxis utility to the system. For instance, value is added to a determination of

susceptibility to thyroid cancer when a medical practitioner can prescribe or administer a standard of care that can reduce susceptibility to the cancer; and/or delay onset of the cancer; and/or increase the likelihood of detecting the cancer at an early stage, to facilitate early treatment when the cancer has not spread and is most curable. Exemplary lifestyle change protocols include loss of weight, increase in exercise, cessation of unhealthy behaviors such as smoking, and change of diet. Exemplary medicinal and surgical intervention protocols include administration of pharmaceutical agents for prophylaxis; and surgery, including in extreme cases surgery to remove a tissue or organ before it has become cancerous. Exemplary diagnostic protocols include non-invasive and invasive imaging; monitoring metabolic biomarkers; and biopsy screening.

For example, in some variations, the system further includes a medical protocol database 214 operatively connected to a computer-readable medium of the system and containing information correlating the presence or absence of the at least one RALGDS allele of interest and medical protocols for human subjects at risk for the cancer. Such medical protocols include any variety of medicines, lifestyle changes, diagnostic tests, increased frequencies of diagnostic tests, and the like that are designed to achieve one of the aforementioned goals. The information correlating a RALGDS allele with protocols could include, for example, information about the success with which the cancer is avoided or delayed, or success with which the cancer is detected early and treated, if a subject has a RALGDS susceptibility allele and follows a protocol. The system of this embodiment further includes a medical protocol tool or routine 216, operatively connected to the medical protocol database 214 and to the analysis tool or routine 210. The medical protocol tool or routine 216 preferably is stored on a computer-readable medium of the system, and adapted to be executed on a processor of the system, to: (i) compare (or correlate) the conclusion that is obtained from the analysis routine 210 (with respect to susceptibility to thyroid cancer for the subject) and the medical protocol database 214, and (ii) generate a protocol report with respect to the probability that one or more medical protocols in the medical protocol database will achieve one or more of the goals of reducing susceptibility to thyroid cancer; delaying onset of thyroid cancer; and increasing the likelihood of detecting thyroid cancer at an early stage to facilitate early treatment. The probability can be based on empirical evidence collected from a population of humans and expressed either in absolute terms (e.g., compared to making no intervention), or expressed in relative terms, to highlight the comparative or additive benefits of two or more protocols.

Some variations of the system just described include the communication tool 212. In some examples, the communication tool generates a communication that includes the protocol report in addition to, or instead of, the conclusion with respect to susceptibility.

Information about RALGDS allele status not only can provide useful information about identifying or quantifying susceptibility to thyroid cancer; it can also provide useful information about possible causative factors for a human subject identified with thyroid cancer, and useful information about therapies for the cancer patient. In some variations, systems of the invention are useful for these purposes.

For instance, in some variations the invention is a system for assessing or selecting a treatment protocol for a subject diagnosed with thyroid cancer. An exemplary system, schematically depicted in Figure 3, comprises:

(a) at least one processor;

(b) at least one computer-readable medium;

(c) a medical treatment database 308 operatively connected to a computer-readable medium of the system and containing information correlating the presence or absence of at least one RALGDS allele and efficacy of treatment regimens for the cancer;

(d) a measurement tool 306 to receive an input (304, depicted in Fig. 3 but not part of the system per se) about the human subject and generate information from the input 304 about the presence or absence of the at least one RALGDS allele indicative of a RALGDS defect in a human subject diagnosed with thyroid cancer; and

(e) a medical protocol routine or tool 310 operatively coupled to the medical treatment database 308 and the measurement tool 306, stored on a computer-readable medium of the system, and adapted to be executed on a processor of the system, to compare the information with respect to presence or absence of the at least one RALGDS allele for the subject and the medical treatment database, and generate a conclusion with respect to at least one of:

(i) the probability that one or more medical treatments will be efficacious for treatment of thyroid cancer for the patient; and (ii) which of two or more medical treatments for thyroid cancer will be more efficacious for the patient.

Preferably, such a system further includes a communication tool 312 operatively connected to the medical protocol tool or routine 310 for communicating the conclusion to the subject 300, or to a medical practitioner for the subject 302 (both depicted in the schematic of Fig. 3, but not part of the system per se). An exemplary communication tool comprises a routine stored on a computer-readable medium of the system and adapted to be executed on a processor of the system, to generate a communication containing the conclusion; and transmit the

communication to the subject or the medical practitioner, or enable the subject or medical practitioner to access the communication.

It will be readily apparent to the skilled person, that the foregoing aspects may be implemented using any of the variants described herein to be associated with risk of thyroid cancer. In certain embodiments, RALGDS alleles are selected from the group consisting of the G allele of rs34170541, the T allele of rsl l3532379, the C allele of rsl39082000, the C allele of chr9: 134971204, the C allele of rsl40573248, and the A allele of chr9: 134977276. In a preferred embodiment, the RALGDS allele is the T allele of rsl 13532379.

The present invention will now be exemplified by the following non-limiting examples.

EXAMPLE 1

High capacity DNA sequencing techniques were used to sequence the entire genomes of about 1187 Icelanders to an average depth of over 10X. This identified some 22 million SNPs. Using imputation assisted by long-range haplotype phasing, sequence data was used to determine the genotypes of the 22 million SNPs in the 59,689 Icelanders who had been genotyped on the SNP chips.

Moreover, knowledge of the Icelandic genealogy allowed for propagation of genotypic information into individuals for whom neither SNP chip nor sequence data were available, a process referred to as "genealogy-based in silico genotyping". Reference is made to the combined method of imputing sequence-derived data into phased chromosomes from chip-typed individuals and using genealogy-based in silico genotyping to infer the sequence of un-genotyped individuals as "two-way imputation". Using this methodology, genotypes for up to about 300,000 individuals may be imputed. The total number of cases entered into this process was 546 individuals with Thyroid cancer.

A two-way imputation-based genome-wide association analysis of the 22 million SNPs was conducted. The analysis revealed strong association of marker rsl 13532379 located on chromosome 9q34.2 with thyroid cancer. The allele specific odds ratios (ORs) of allele T of this variant is 3.00, with a P-value of 2.3 xlO^"04 thus representing a novel risk variant for thyroid cancer. Furthermore, rsl l3532379 is located in the protein coded region of the RALGDS gene where the base change results in an amino acid change from Glycine to Serine at position 713 in the protein i.e. G713S. Replication study on Dutch and Spanish case:control groups confirms the association to this marker with combined P-value of 4.3xl0^"03,

A follow-up study of this finding was conducted by Single SNP genotyping 548 Icelandic cases vs 3564 controls using Centaurus assay (Nanogen, Kutyavin, et al., (2006), Nucleic Acids Res, 34, el28). Table 2 summarizes results obtained for rsll3532379 on chromosome 9q34.2.

Table 2. Results of association of rsl l3532379 on chromosome 9 with Thyroid Cancer. Shown are P-values of association (after correction for relatedness; see Methods (Example 4)), OR of the risk allele T, and where applicable; number of affected, risk allele frequency in affected, number of controls, risk allele frequency in controls.

EXAMPLE 2

A number of missense variants in the RALGDS gene were investigated for association with thyroid cancer, using data based on imputation methods, as described in the above, and in more detail under Example 4.

Results of this analysis are presented in Table 3. As can be seen, several missense variants were found within the RALGDS gene. Marker rsl39082000 in particular shows evidence (OR 2.16) for association with thyroid cancer, in addition to rsl 13532379. The rsl39082000 marker encodes a serine to cysteine substitution at position 686 in the RALGDS gene.

Table 3. Association results for missense variants in the RALGDS gene. Shown are P-values of association, OR for minor allele, marker identity, minor allele frequency (in percentages), position on chromosome 9 in NCBI Build 36, identity of minor and major alleles, identity of the coding change encoded by the polymorphism, position of coding change in RALGDS protein, and reference to Seq ID No for flanking sequence of the polymorphism.

EXAMPLE 3

Follow-up analysis of genome-wide sequencing data obtained as described in the above under Example 1, using experimental methods described in the below (Example 4), a number of further variants were identified that associate with risk of thyroid cancer.

These variants are identified in Table 4, together with results of their association with thyroid cancer.

Table 4. Association results for markers with thyroid cancer. Shown are marker names,

chromosomal location and position in NCBI Build 36, P-value of association with thyroid cancer, odds ratio, risk allele frequency, identity of risk allele and other allele for the SNP, and seq id of flanking sequence of the SNP.

EXAMPLE 4

Methods

Subjects

ICELAND. Approval for this study was granted by the National Bioethics Committee of Iceland and the Icelandic Data Protection Authority.

The collection of samples used for the thyroid cancer study represents the overall distribution in Iceland quite well. Of the cases that genotypes were generated for, either by directly genotyping or in-silico genotyping, about 80% are of papillary type, about 12% are of follicular type, about 2% are medullary thyroid cancer, and the remainders are of unknown or undetermined histological sub-phenotype. The results presented in Table 2 are for the combined results for all thyroid cases since no statistically significant difference was observed between the different histological subgroups.

The Icelandic controls consist of about 60,000 individuals from other ongoing genome-wide association studies at deCODE genetics. Individuals with a diagnosis of thyroid cancer were excluded. Both male and female genders were included .

SPAIN. The Spanish study population consisted of 90 non-medullary thyroid cancer cases. The cases were recruited from the Oncology Department of Zaragoza Hospital in Zaragoza, Spain, from October 2006 to June 2007. All patients were of self-reported European descent. Clinical information including age at onset, grade and stage was obtained from medical records. The average age at diagnosis for the patients was 48 years (median 49 years) and the range was from 22 to 79 years. The 1,399 Spanish control individuals 798 (57%) males and 601 (43%) females had a mean age of 51 (median age 50 and range 12-87 years) were approached at the University Hospital in Zaragoza, Spain, and were not known to have thyroid cancer. The DNA for both the Spanish cases and controls was isolated from whole blood using standard methods. Study protocols were approved by the Institutional Review Board of Zaragoza University

Hospital. All subjects gave written informed consent.

THE NETHERLANDS. The Dutch study population consists of 151 non-medullary thyroid cancer cases (75% are females) and 832 cancer-free individuals (54% females). The cases were recruited from the Department of Endocrinology, Radboud University Nijmegen Medical Centre (RUNMC), Nijmegen, The Netherlands from November 2009 to June 2010. All patients were of self-reported European descent. Demographic, clinical, tumor treatment and follow-up related characteristics were obtained from the patient's medical records. The average age at diagnosis for the patients was 39 years (SD 12.8). The DNA for both the Dutch cases and controls was isolated from whole blood using standard methods. The controls were recruited within a project entitled "Nijmegen Biomedical Study" (NBS). The details of this study were reported previously (Wetzels, et al.. Kidney Int 72, 632-7 (2007)). Control individuals from the NBS were invited to participate in a study on gene-environment interactions in multifactorial diseases such as cancer. They were all of self-reported European descent and fully informed about the goals and the procedures of the study. The study was approved by the Ethical Committee and the Institutional Review Board of the RUNMC, Nijmegen, The Netherlands and all study subjects gave written informed consent.

Single track assay SNP genotyping. Single SNP genotyping for the two case-control groups from Iceland The Netherlands and Spain was carried out by deCODE Genetics in Reykjavik, Iceland, applying the Centaurus (Nanogen) platform (Kutyavin, IV et al Nucleic Acids Res 34:el28 (2006)). The quality of each Centaurus SNP assay was evaluated by genotyping each assay in the CEU and/or YRI HapMap samples and comparing the results with the HapMap publicly released data. Assays with > 1.5% mismatch rate were not used and a linkage disequilibrium

(LD) test was used for markers known to be in LD. We genotyped 330 individuals using both the Illumina Hap300 chip and Centaurus single track SNP assay and observed a mismatch rate lower than 0.5%.

Illumina SNP Chip Genotyping: The Icelandic chip-typed samples were assayed with the Illumina Human Hap300, Hap CNV370, Hap 610, 1M or Omni-1 Quad bead chips at deCODE genetics. Only the 317,503 SNPs from the Human Hap300 chip were used in the long range phasing and the subsequent SNP imputations. SNPs were excluded if they had (i) yield lower than 95%, (ii) minor allele frequency less than 1% in the population or (iii) significant deviation from Hardy- Weinberg equilibrium in the controls (P < 0.001), (iv) if they produced an excessive inheritance error rate (over 0.001), (v) if there was substantial difference in allele frequency between chip types (from just a single chip if the problem that resolved all differences, but from all chips otherwise). All samples with a call rate below 97% were excluded from the analysis. The final set of SNPs used for long range phasing was composed of 297,835 autosomal SNPs.

Whole Genome Sequencing and SNP Calling: SNPs were imputed based on whole genome sequence data from about 1176 Icelanders, selected for various neoplastic, cardiovascular and psychiatric conditions. All of the individuals were sequenced at a depth of at least 10X.

Approximately 22 million SNPs were imputed based on this set of individuals.

a. Sample preparation. Paired-end libraries for sequencing were prepared according to the manufacturer's instructions (Illumina). In short, approximately 5 μg of genomic DNA, isolated from frozen blood samples, was fragmented to a mean target size of 300 bp using a Covaris E210 instrument. The resulting fragmented DNA was end repaired using T4 and Klenow polymerases and T4 polynucleotide kinase with 10 mM dNTP followed by addition of an 'A' base at the ends using Klenow exo fragment (3' to 5'-exo minus) and dATP (1 mM). Sequencing adaptors containing overhangs were ligated to the DNA products followed by agarose (2%) gel electrophoresis. Fragments of about 400 bp were isolated from the gels (QIAGEN Gel

Extraction Kit), and the adaptor-modified DNA fragments were PCR enriched for ten cycles using Phusion DNA polymerase (Finnzymes Oy) and PCR primers PE 1.0 and PE 2.0 (Illumina).

Enriched libraries were further purified using agarose (2%) gel electrophoresis as described above. The quality and concentration of the libraries were assessed with the Agilent 2100 Bioanalyzer using the DNA 1000 LabChip (Agilent). Barcoded libraries were stored at -20 °C. All steps in the workflow were monitored using an in-house laboratory information management system with barcode tracking of all samples and reagents.

b. DNA sequencing. Template DNA fragments were hybridized to the surface of flow cells (Illumina PE flowcell, v4) and amplified to form clusters using the Illumina cBot. In brief, DNA (8-10 pM) was denatured, followed by hybridization to grafted adaptors on the flowcell.

Isothermal bridge amplification using Phusion polymerase was then followed by linearization of the bridged DNA, denaturation, blocking of 3 ^' ends and hybridization of the sequencing primer. Sequencing-by-synthesis was performed on Illumina GAIIx instruments equipped with paired- end modules. Paired-end libraries were sequenced using 2 x 101 cycles of incorporation and imaging with Illumina sequencing kits, v4 or v5 (TruSeq) . Each library or sample was initially run on a single lane for validation followed by further sequencing of ≥4 lanes with targeted raw cluster densities of 500-700 k/mm², depending on the version of the data imaging and analysis packages. Imaging and analysis of the data was performed using either the SCS2.6 /RTA1.6 or SCS2.8/RTA1.8 software packages from Illumina, respectively. Real-time analysis involved conversion of image data to base-calling in real-time.

c. Alignment. For each lane in the DNA sequencing output, the resulting qseq files were converted into fastq files using an in-house script. All output from sequencing was converted, and the Illumina quality filtering flag was retained in the output. The fastq files were then aligned against Build 36 of the human reference sequence using bwa version 0.5.7. All genomic locations quoted refer to HG18 Build 36.

d. BAM file generation. SAM file output from the alignment was converted into BAM format using SAMtools version 0.1.8, and an in-house script was used to carry the Illumina quality filter flag over to the BAM file. The BAM files for each sample were then merged into a single BAM file using SAMtools. Finally, Picard version 1.17 (see http://picard.sourceforge.net/) was used to mark duplicates in the resulting sample BAM files.

e. SNP identification and genotype calling: A two-step approach was applied. The first step was to detect SNPs by identifying sequence positions where at least one individual could be determined to be different from the reference sequence with confidence (quality threshold of 20) based on the SNP calling feature of the pileup tool in SAMtools. SNPs that always differed heterozygous or homozygous from the reference were removed. The second step was to use the pileup tool to genotype the SNPs at the positions that were flagged as polymorphic. Because sequencing depth varies and hence the certainty of genotype calls also varies, genotype likelihoods rather than deterministic calls were calculated. Of the 2.5 million SNPs reported in the HapMap2 CEU samples, 96.3% were observed in the whole-genome sequencing data. Of the 6.9 million SNPs reported in the 1000 Genomes Project data, 89.4% were observed in the whole- genome sequencing data.

Genotype Imputation Methods

Long range phasing: Long range phasing of all chip-genotyped individuals was performed with methods described previously (Kong, A. et al. Nat Genet 40, 1068-75 (2008); Holm, H. et al. Nat Genet 43, 316-20 (2011)). In brief, phasing is achieved using an iterative algorithm which phases a single proband at a time given the available phasing information about everyone else that shares a long haplotype identically by state with the proband. Given the large fraction of the Icelandic population that has been chip-typed, accurate long range phasing is available genome- wide for all chip-typed Icelanders. For long range phased haplotype association analysis, we then partitioned the genome into non-overlapping fixed 0.3cM bins. Within each bin, we observed the haplotype diversity described by the combination of all chip-typed markers in the bin. Haplotypes with frequencies over 0.001 were tested in a case: control analysis.

Genotype imputation: We imputed the SNPs identified and genotyped through sequencing into all Icelanders who had been phased with long range phasing using the same model as used by IMPUTE (Kong, A. et al. Nat Genet 40, 1068-75 (2008)). The genotype data from sequencing can be ambiguous due to low sequencing coverage. In order to phase the sequencing genotypes, an iterative algorithm was applied for each SNP with alleles 0 and 1. We let H be the long range phased haplotypes of the sequenced individuals and applied the following algorithm:

1. For each haplotype h in H, use the Hidden Markov Model of IMPUTE to calculate for every other k in H, the likelihood, denoted y_hrk, of h having the same ancestral source as k at the SNP.

2. For every h in H, initialize the parameter _h, which specifies how likely the one allele of the SNP is to occur on the background of h from the genotype likelihoods obtained from sequencing. The genotype likelihood L_g is the probability of the observed sequencing data at the SNP for a given individual assuming g is the true genotype at the SNP. If L₀, Li and L₂ are the likelihoods of the genotypes 0, 1 and 2 in the individual that carries h, then set e_h = -^-.

3. For every pair of haplotypes h and k in H that are carried by the same individual, use the other haplotypes in H to predict the genotype of the SNP on the backgrounds of h and k:

Th = \{h}Yh and r_k =∑i_€H\{k)Yk,i9i- Combining these predictions with the genotype likelihoods from sequencing gives un-normalized updated phased genotype probabilities:

Λ>ο = (1 - τ„)(ΐ - T_k)L_0l P₁₀ = τ„(ΐ - P₀₁ = (l - and p_u = T_hr_kL₂. Now use these values to update 9_h and _k to 0_h =— — and 0_k =

4. Repeat step 3 when the maximum difference between iterations is greater than a

convergence threshold ε. We used ε=10^"7.

Given the long range phased haplotypes and Θ, the allele of the SNP on a new haplotype h not in H, is imputed as∑_l€Hr_hA-

The above algorithm can easily be extended to handle simple family structures such as parent- offspring pairs and triads by letting the P distribution run over all founder haplotypes in the family structure. The algorithm also extends trivially to the X-chromosome. If source genotype data are only ambiguous in phase, such as chip genotype data, then the algorithm is still applied, but all but one of the Ls will be 0. In some instances, the reference set was intentionally enriched for carriers of the minor allele of a rare SNP in order to improve imputation accuracy. In this case, expected allele counts will be biased toward the minor allele of the SNP. Call the enrichment of the minor allele and let 9' be the expected minor allele count calculated from the naive imputation method, and let Θ be the unbiased expected allele count, then θ = _{θ θ} and hence θ =— -— .

Ε+(1-Ε)θ'

This adjustment was applied to all imputations based on enriched imputations sets. We note that if θ' is 0 or 1, then Θ will also be 0 or 1, respectively.

Genotype imputation information: The informativeness of genotype imputation was estimated by the ratio of the variance of imputed expected allele counts and the variance of the actual allele counts:

Var(E(6 \chip data))

Var {0) '

where Θ e {0, 1} is the allele count. Var(E(0 \chip data)) was estimated by the observed variance of the imputed expected counts and Var{6) was estimated by p(i - p), where p is the allele frequency. The information value for chr9:94276933 was 0.99.

Genealogy-based in silico genotyping: In addition to imputing sequence variants from the whole genome sequencing effort into chip genotyped individuals, we also performed a second imputation step where genotypes were imputed into relatives of chip genotyped individuals, creating in silico genotypes. The inputs into the second imputation step are the fully phased (in particular every allele has been assigned its parent of origin (Kong, A. et al. Nature 462, 868-74 (2009)) imputed and chip type genotypes of the available chip typed individual. The algorithm used to perform the second imputation step consists of:

1. For each ungenotyped individual (the proband), find all chip genotyped individuals within two meiosis of the individual. The six possible types of two meiotic distance relatives of the proband are (ignoring more complicated relationships due to pedigree loops) :

Parents, full and half siblings, grandparents, children and grandchildren. If all pedigree paths from the proband to a genotyped relative go through other genotyped relatives, then that relative is excluded. E.g. if a parent of the proband is genotyped, then the proband's grandparents through that parent are excluded. If the number of meiosis in the pedigree around the proband exceeds a threshold (we used 12), then relatives are removed from the pedigree until the number of meiosis falls below 12, in order to reduce computational complexity.

2. At every point in the genome, calculate the probability of each genotyped relative sharing with the proband based on the autosomal SNPs used for phasing. A multipoint algorithm based on the hidden Markov model Lander-Green multipoint linkage algorithm using fast Fourier transforms is used to calculate these sharing probabilities (Lander, E.S. & Green, P. Proc Natl Acad Sci U S A 84, 2363-7 (1987); Kruglyak, L. & Lander, E.S. J Comput Biol 5, 1-7 (1998)) . First single point sharing probabilities are calculated by dividing the genome into 0.5cM bins and using the haplotypes over these bins as alleles. Haplotypes that are the same, except at most at a single SNP, are treated as identical. When the haplotypes in the pedigree are incompatible over a bin, then a uniform probability distribution was used for that bin. The most common causes for such incompatibilities are recombinations within the pedigree, phasing errors and genotyping errors. Note that since the input genotypes are fully phased, the single point information is substantially more informative than for unphased genotyped, in particular one haplotype of the parent of a genotyped child is always known. The single point distributions are then convolved using the multipoint algorithm to obtain multipoint sharing probabilities at the center of each bin. Genetic distances were obtained from the most recent version of the deCODE genetic map (Kong, A. et al. Nature 467, 1099-103 (2010)).

Based on the sharing probabilities at the center of each bin, all the SNPs from the whole genome sequencing are imputed into the proband. To impute the genotype of the paternal allele of a SNP located at x, flanked by bins with centers at x_left and χ_τί9Μ ·

Starting with the left bin, going through all possible sharing patterns v, let l_v be the set of haplotypes of genotyped individuals that share identically by descent within the pedigree with the proband's paternal haplotype given the sharing pattern v and P y) be the probability of v at the left bin - this is the output from step 2 above - and let e_; be the expected allele count of the SNP for haplotype i . Then e_v = ^⁶'"⁶' is the expected allele

∑ie/_v 1

count of the paternal haplotype of the proband given v and an overall estimate of the allele count given the sharing distribution at the left bin is obtained from e_left =∑_v P(v)e_v. If l_v is empty then no relative shares with the proband's paternal haplotype given v and thus there is no information about the allele count. We therefore store the probability that some genotyped relative shared the proband's paternal haplotype, O_left =∑_ν,ι_ν=φ Ρ(ν) and an expected allele count, conditional on the proband's paternal haplotype being shared by at least one genotyped relative: c_left = ^*ⁱ—— in the same way calculate

O_right and c_right . Linear interpolation is then used to get an estimates at the SNP from the two flanking bins:

o = o_left + - — O_right - O_left),

right ^xleft

x^{~ x}left

^eft ^ _ _v V-rlght Heft)

Xright ^xleft

If Θ is an estimate of the population frequency of the SNP then Oc + (1 - 0)θ is an estimate of the allele count for the proband's paternal haplotype. Similarly, an expected allele count can be obtained for the proband's maternal haplotype.

Case: control association testing: Logistic regression was used to test for association between SNPs and disease, treating disease status as the response and expected genotype counts from imputation or allele counts from direct genotyping as covariates. Testing was performed using the likelihood ratio statistic. When testing for association using the in silico genotypes, controls were matched to cases based on the informativeness of the imputed genotypes, such that for each case C controls of matching informativeness where chosen. Failing to match cases and controls will lead to a highly inflated genomic control factor, and in some cases may lead to spurious false positive findings. The informativeness of each of the imputation of each one of an individual's haplotypes was estimated by taki average of

over all SNPs imputed for the individual, where e is the expected allele count for the haplotype at the SNP and Θ is the population frequency of the SNP. Note that α(θ, θ) = 0 and α(θ, β) = α(ΐ, θ) = 1. The mean informativeness values cluster into groups corresponding to the most common pedigree configurations used in the imputation, such as imputing from parent into child or from child into parent. Based on this clustering of imputation informativeness we divided the haplotypes of individuals into seven groups of varying informativeness, which created 27 groups of individuals of similar imputation informativeness; 7 groups of individuals with both haplotypes having similar informativeness, 21 groups of individuals with the two haplotypes having different informativeness, minus the one group of individuals with neither haplotype being imputed well. Within each group we calculate the ratio of the number of controls and the number of cases, and choose the largest integer C that was less than this ratio in all the groups. For example, if in one group there are 10.3 times as many controls as cases and if in all other groups this ratio was greater, then we would set C = 10 and within each group randomly select ten times as many controls as there are cases.

Inflation Factor Adjustment: In order to account for the relatedness and stratification within the case and control sample sets we applied the method of genomic control based on chip typed markers. Quoted P values were adjusted accordingly.

Effective sample size estimation: In order to estimate the effective sample size of the case control association analyses, we compared the variances of the logistic and generalized linear regression parameter estimates based on the in silico genotypes to their one step imputation counterparts. For the quantitative trait association analysis, assume that a single step imputation (SNPs are imputed, but in silico genotypes are not used) association analysis with n_x subjects leads on average to an estimate of the regression parameter with variance

and that the corresponding in silico genotype association analysis leads to an estimate of the regression parameter with variance σ\, then assuming that variance goes down linearly with sample size we estimate the effective sample size in the in silico genotype association analysis as n₂ = ¾n_x. For the case control association analysis, the number of controls is much greater than the number of cases and we use the same formula to estimate the effective number of cases, with the n-s representing the number of cases and the a²-s representing the variances of the logistic regression coefficient.

Claims

1. A method of determining a susceptibility to thyroid cancer, the method comprising : analyzing data about at least one allele of human RALGDS gene (SEQ ID NO: 57) in a human subject, wherein different alleles of the human RALGDS gene are associated with different susceptibilities to thyroid cancer in humans, and determining a susceptibility to thyroid cancer for the human subject from the data.

2. The method of claim 1, comprising analyzing the data for the presence or absence of at least one mutant allele indicative of a RALGDS defect selected from the group consisting of:

(a) an encoded RALGDS protein with an altered amino acid sequence, relative to the RALGDS amino acid sequence set forth in SEQ ID NO: 58;

(b) expression of a RALGDS protein with reduced activity compared to a wild-type RALGDS protein (SEQ ID NO: 58), wherein the activity is at least one RALGDS activity selected from :

(i) guanine nucleotide exchange activity; and

(ii) RAS binding activity

(c) reduced expression of RALGDS protein, compared to wild-type RALGDS, wherein mutant alleles indicative of the defect are associated with increased susceptibility to thyroid cancer.

3. The method of claim 1 or claim 2, comprising analyzing the data for the presence or absence of at least one mutant allele that results in elimination of the at least one activity.

4. The method of any one of the previous claims, wherein the analyzing data comprises analyzing a biological sample from the human subject to obtain information selected from the group consisting of:

(a) nucleic acid sequence information, wherein the nucleic acid sequence information comprises sequence sufficient to identify the presence or absence of the mutant allele in the subject; (b) nucleic acid sequence information, wherein the nucleic acid sequence information identifies at least one allele of a polymorphic marker in linkage disequilibrium (LD) with the mutant allele, wherein the LD is characterized by a value for r² of at least 0.5;

(c) measurement of the quantity or length of RALGDS mRNA, wherein the measurement is indicative of the presence or absence of the mutant allele;

(d) measurement of the quantity of RALGDS protein, wherein the measurement is indicative of the presence or absence of the mutant allele; and

(e) measurement of RALGDS activity, wherein the measurement is indicative of the presence or absence of the mutant allele.

5. The method of claim 4, comprising analyzing the biological sample to obtain the nucleic acid sequence information.

6. The method of claim 4 or claim 5, further comprising obtaining a biological sample comprising nucleic acid from the human subject.

7. The method of any one of claims 1 - 3, wherein the analyzing data comprises analyzing data from a preexisting record about the human subject.

8. The method of any one of the preceding claims, wherein the presence of the allele is indicative of increased susceptibility to thyroid cancer with a relative risk (RR) or odds ratio (OR) of at least 1.5, of at least 1.6, of at least 1.7, of at least 1.8, at least 1.9, of at least 2.0, at least 2.5, of at least 3.0, at least 3.5, of at least 4.0, at least 4.5, of at least 5.0, at least 5.5, of at least 6.0, at least 7.0, or of at least 8.0.

9. The method of any one of the preceding claims, wherein the at least one allele is selected from the group consisting of rs34170541 allele G, rsl l3532379 allele T, rsl39082000 allele C, chr9: 134971204 allele C, rsl40573248 allele C, and chr9: 134977276 allele A.

10. The method of any one of the claims 1 to 8, wherein the at least one allele is the T allele of marker rsl l3532379, or a marker allele in linkage disequilibrium therewith.

11. The method of claim 10, wherein the marker allele in linkage disequilibrium with the T allele of marker rsl l3532379 is selected from the group consisting of the marker alleles set forth in Table 1.

12. The method of claim 1, wherein the allele comprises a RALGDS missense mutation.

13. The method of claim 12, wherein the missense mutation is selected from the group consisting of an isoleucine to leucine substitution at position 724 in RALGDS protein (SEQ ID NO: 58), a glycine to serine substitution at position 713 in RALGDS protein (SEQ ID NO: 58), a serine to cysteine substitution at position 686 in RALGDS protein (SEQ ID NO: 58), a glutamine to arginine substitution at position 513 in RALGDS protein (SEQ ID NO: 58), a lycine to arginine substitution at position 222 in RALGDS protein (SEQ ID NO:58), and a glycine to cysteine substitution at position 90 in RALGDS protein (SEQ ID NO: 58) .

14. The method of claim 13, wherein the missense mutation is a glycine to serine substitution at position 713 in RALGDS protein (SEQ ID NO: 58) .

15. The method of claim 1, wherein the mutant allele is a RALGDS nonsense mutation.

16. The method of claim 12, wherein the mutant allele is a RALGDS missense mutation which results in expression of a RALGDS protein with reduced activity compared to a wild-type RALGDS protein.

17. The method of claim 16, wherein the missense mutation results in elimination of RALGDS activity.

18. A method of determining whether a human subject is at increased risk of developing thyroid cancer, the method comprising steps of obtaining a biological sample containing nucleic acid from the subject; determining, in the biological sample, nucleic acid sequence about the human RALGDS gene; and comparing the sequence information to the wild-type sequence of RALGDS (SEQ ID NO: 57); wherein an identification of a mutation in RALGDS in the individual is indicative that the subject is at increased risk of developing thyroid cancer.

19. The method of claim 18, wherein the mutation encodes a missense mutation, a nonsense mutation or a frameshift mutation in RALGDS with sequence as set forth in SEQ ID NO: 58.

20. The method of claim 18 or claim 19, wherein the mutation results in a RALGDS defect selected from the group consisting of:

(a) an encoded RALGDS protein with an altered amino acid sequence, relative to the RALGDS amino acid sequence set forth in SEQ ID NO: 58; (b) expression of a RALGDS protein with reduced activity compared to a wild-type RALGDS protein (SEQ ID NO: 58), wherein the activity is at least one RALGDS activity selected from :

(i) guanine nucleotide exchange activity; and

(ii) RAS binding activity;

21. A method of determining whether a human subject is at increased risk of developing thyroid cancer, the method comprising analyzing amino acid sequence data about a RALGDS polypeptide from the subject, wherein a determination of the presence of a RALGDS polypeptide with altered sequence compared with a wild-type RALGDS polypeptide with sequence as set forth in SEQ ID NO: 58 is indicative that the subject is at increased risk of developing thyroid cancer.

22. The method of claim 21, wherein the amino acid sequence data is obtained from a biological sample from the human subject comprising human RALGDS polypeptide, using a method that comprises at least one procedure selected from :

(i) an antibody assay; and

(ii) protein sequencing.

23. The method of claim 21, wherein the amino acid sequence data is obtained from a preexisting record.

24. A method for determining a susceptibility to thyroid cancer in a human subject, comprising determining the presence or absence of at least one allele of at least one polymorphic marker in a nucleic acid sample obtained from the subject, wherein the at least one polymorphic marker is selected from the group consisting of

(a) polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58; and (b) rsl2407041, rs6689698, rsl2136101, rsl 1586476, rs62174266, rs76839330, rsl354833, rs75825480, rs6828277, rsl7540362, rsl060412, rs55635625, rs497341, rs73205431, rs8021657, rsl46663071, rs28524987 and rs62223910; and determining a susceptibility to thyroid cancer from the presence or absence of the at least one allele, wherein the presence of the at least one allele is indicative of a susceptibility to thyroid cancer for the subject.

25. The method of claim 24, wherein the polymorphic markers in the human RALGDS gene are selected from the group consisting of polymorphic markers that encode a missense mutation or a nonsense mutation in RALGDS with sequence as set forth in SEQ ID NO: 58.

26. The method of claim 24, wherein the polymorphic markers in the human RALGDS gene are selected from the group consisting of rs34170541, rsl 13532379, rsl39082000,

chr9: 134971204, rsl40573248 and chr9: 134977276.

27. The method of any one of the preceding claims, wherein the subject has previously not been diagnosed with thyroid cancer.

28. An assay for determining a susceptibility to thyroid cancer in a human subject, the assay comprising steps of:

(i) obtaining a nucleic acid sample from the human subject

(ii) assaying the nucleic acid sample to determine the presence or absence of at least one allele of at least one polymorphic marker associated with increased susceptibility to thyroid cancer in humans, and

(iii) determining a susceptibility to thyroid cancer for the human subject from the presence or absence of the at least one allele, wherein the at least one polymorphic marker is selected from the group consisting of

(a) polymorphic markers in the human RALGDS gene that result in an encoded RALGDS with an altered amino acid sequence compared with wild-type RALGDS with amino acid sequence as set forth in SEQ ID NO: 58; and

(b) rsl2407041, rs6689698, rsl2136101, rsl 1586476, rs62174266, rs76839330, rsl354833, rs75825480, rs6828277, rsl7540362, rsl060412, rs55635625, rs497341, rs73205431, rs8021657, rsl46663071, rs28524987 and rs62223910; and wherein a susceptibility to thyroid cancer is determined from the presence or absence of the at least one allele.

29. A system for identifying susceptibility to thyroid cancer in a human subject, the system comprising : at least one processor; at least one computer-readable medium; a susceptibility database operatively coupled to a computer-readable medium of the system and containing population information correlating the presence or absence of one or more alleles of the human RALGDS gene and susceptibility to thyroid cancer in a population of humans; a measurement tool that receives an input about the human subject and generates information from the input about the presence or absence of at least one mutant RALGDS allele indicative of a RALGDS defect in the human subject; and an analysis tool that: is operatively coupled to the susceptibility database and the the measurement tool, is stored on a computer-readable medium of the system, is adapted to be executed on a processor of the system, to compare the information about the human subject with the population information in the susceptibility database and generate a conclusion with respect to susceptibility to thyroid cancer for the human subject.

30. The system according to claims 29, further including : a communication tool operatively coupled to the analysis tool, stored on a computer-readable medium of the system and adapted to be executed on a processor of the system to communicate to the subject, or to a medical practitioner for the subject, the conclusion with respect to susceptibility to thyroid cancer for the subject.

31. The system according to claim 29 or claim 30, wherein the at least one mutant RALGDS allele is indicative of a RALGDS defect selected from the group consisting of:

(i) guanine nucleotide exchange activity; and

(ii) RAS binding activity; (c) reduced expression of RALGDS protein, compared to wildtype expression, wherein mutant alleles indicative of the defect are associated with increased susceptibility to thyroid cancer.

32. The system according to any one of the claims 29 to 31, wherein the measurement tool comprises a tool stored on a computer-readable medium of the system and adapted to be executed by a processor of the system to receive a data input about a subject and determine information about the presence or absence of the at least one mutant RALGDS allele in a human subject from the data.

33. The system according to claim 32, wherein the data is genomic sequence information, and the measurement tool comprises a sequence analysis tool stored on a computer readable medium of the system and adapted to be executed by a processor of the system to determine the presence or absence of the at least one mutant RALGDS allele from the genomic sequence information.

34. The system according to any one of claims 29-33, wherein the input about the human subject is a biological sample from the human subject, and wherein the measurement tool comprises a tool to identify the presence or absence of the at least one mutant RALGDS allele in the biological sample, thereby generating information about the presence or absence of the at least one mutant RALGDS allele in a human subject.

35. The system according to claim 34, wherein the measurement tool includes: an oligonucleotide microarray containing a plurality of oligonucleotide probes attached to a solid support; a detector for measuring interaction between nucleic acid obtained from or amplified from the biological sample and one or more oligonucleotides on the oligonucleotide microarray to generate detection data; and an analysis tool stored on a computer-readable medium of the system and adapted to be executed on a processor of the system, to determine the presence or absence of the at least one mutant RALGDS allele based on the detection data.

36. The system according to claim 34, wherein the measurement tool includes: a nucleotide sequencer capable of determining nucleotide sequence information from nucleic acid obtained from or amplified from the biological sample; and an analysis tool stored on a computer-readable medium of the system and adapted to be executed on a processor of the system, to determine the presence or absence of the at least one mutant RALGDS allele based on the nucleotide sequence information.

37. The system according to any one of claims 29 to 36, further comprising : a medical protocol database operatively connected to a computer-readable medium of the system and containing information correlating the presence or absence of the at least one mutant RALGDS allele and medical protocols for human subjects at risk for thyroid cancer; and a medical protocol routine, operatively connected to the medical protocol database and the analysis routine, stored on a computer-readable medium of the system, and adapted to be executed on a processor of the system, to compare the conclusion from the analysis routine with respect to susceptibility to thyroid cancer for the subject and the medical protocol database, and generate a protocol report with respect to the probability that one or more medical protocols in the database will : reduce susceptibility to thyroid cancer; or delay onset of thyroid cancer; or increase the likelihood of detecting thyroid cancer at an early stage to facilitate early treatment.

38. The system according to any one of claims 30-37, wherein the communication tool is operatively connected to the analysis routine and comprises a routine stored on a computer- readable medium of the system and adapted to be executed on a processor of the system, to: generate a communication containing the conclusion; and transmit the communication to the subject or the medical practitioner, or enable the subject or medical practitioner to access the communication.

39. The system according to claim 38, wherein the communication expresses the susceptibility to thyroid cancer in terms of odds ratio or relative risk or lifetime risk.

40. The system according to claim 38 or claim 39, wherein the communication further includes the protocol report.

41. The system according to any one of claims 29-40, wherein the susceptibility database further includes information about at least one parameter selected from the group consisting of age, sex, ethnicity, race, medical history, weight, blood pressure, family history of thyroid cancer, and smoking history in humans and impact of the at least one parameter on

susceptibility to thyroid cancer.

42. A system for assessing or selecting a treatment protocol for a subject diagnosed with thyroid cancer, comprising : at least one processor; at least one computer-readable medium; a medical treatment database operatively connected to a computer-readable medium of the system and containing information correlating the presence or absence of at least one mutant RALGDS allele and efficacy of treatment regimens for thyroid cancer; a measurement tool to receive an input about the human subject and generate information from the input about the presence or absence of the at least one mutant RALGDS allele indicative of a RALGDS defect in a human subject diagnosed with thyroid cancer; and a medical protocol tool operatively coupled to the medical treatment database and the measurement tool, stored on a computer-readable medium of the system, and adapted to be executed on a processor of the system, to compare the information with respect to presence or absence of the at least one mutant RALGDS allele for the subject and the medical treatment database, and generate a conclusion with respect to at least one of: the probability that one or more medical treatments will be efficacious for treatment of thyroid cancer for the patient; and which of two or more medical treatments for thyroid cancer will be more efficacious for the patient.

43. The system according to claim 42, wherein the measurement tool comprises a tool stored on a computer-readable medium of the system and adapted to be executed by a processor of the system to receive a data input about a subject and determine information about the presence or absence of the at least one mutant RALGDS allele in a human subject from the data.

44. The system according to claim 43, wherein the data is genomic sequence information, and the measurement tool comprises a sequence analysis tool stored on a computer readable medium of the system and adapted to be executed by a processor of the system to determine the presence or absence of the at least one mutant RALGDS allele from the genomic sequence information.

45. The system according to claim 42, wherein the input about the human subject is a biological sample from the human subject, and wherein the measurement tool comprises a tool to identify the presence or absence of the at least one mutant RALGDS allele in the biological sample, thereby generating information about the presence or absence of the at least one mutant RALGDS allele in a human subject.

46. The system according to any one of claims 42-45, further comprising a communication tool operatively connected to the medical protocol routine for communicating the conclusion to the subject, or to a medical practitioner for the subject.

47. The system according to claim 46, wherein the communication tool comprises a routine stored on a computer-readable medium of the system and adapted to be executed on a processor of the system, to: generate a communication containing the conclusion; and transmit the communication to the subject or the medical practitioner, or enable the subject or medical practitioner to access the communication.