WO2015011262A1

WO2015011262A1 - In vitro genetic diagnostic of inherited peripheral neuropathies (charcot-marie-tooth disease)

Info

Publication number: WO2015011262A1
Application number: PCT/EP2014/066025
Authority: WO
Inventors: Valérie DELAGUE; Nicolas Levy; Patrice Bourgeois; Sylvain BAULANDE
Original assignee: INSERM (Institut National de la Santé et de la Recherche Médicale); Université D'aix-Marseille; Assistance Publique Hôpitaux De Marseille
Priority date: 2013-07-26
Filing date: 2014-07-25
Publication date: 2015-01-29

Abstract

A method of identifying in vitro molecular causes of Charcot Marie Tooth disease, comprising the following steps: (i) providing a physiological sample comprising a genome of a subject, and (ii) implementing on said sample at least one of Process A and Process B, wherein - Process A is determining a number of copy number variation(s), with respect to a sample of a normal subject, on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1; and - Process B is determining a number of point mutation(s), with respect to a sample of a normal subject, on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1.

Description

In vitro genetic diagnostic of Inherited Peripheral Neuropathies (Charcot-

Marie-Tooth disease) FIELD OF THE INVENTION

The present invention relates to an in vitro method, the preparation and the application thereof, for determining, more rapidly, more accurately and at lower cost than the prior art techniques, the molecular basis of Inherited Peripheral Neuropathies (IPNs), mostly represented by Charcot-Marie-Tooth disease (CMT), a family of neuromuscular diseases (NMDs).

BACKGROUND OF THE INVENTION

Inherited Neuromuscular Disorders (hereafter NMDs) form a large and very heterogeneous group of genetic diseases that cause progressive degeneration of the muscles and/or motor nerves that control movements. These pathologies are present in all populations, affecting children as well as adults. The overall prevalence of NMDs is very difficult to evaluate, but one can estimate that, given the incidence of every different type, around 1 out of 1000 people may have a disabling inherited neuromuscular disorder. Most NMD types result in chronic long term disability posing a significant burden to the patients, their families and public health care.

Inherited Peripheral Neuropathies (IPNs), mostly represented by Charcot- Marie-Tooth disease (hereafter CMT), constitute a family of NMDs. Skre, H. (1 ) discloses that CMT is among the most common inherited neurological diseases, with an overall prevalence of about approximately 1 -4/10,000.

Pareyson, D. et al. (2) discloses that it is a genetically heterogeneous group of disorders sharing the same clinical phenotype, characterized by distal limb muscle wasting and weakness, usually with skeletal deformities, distal sensory loss, and abnormalities of deep tendon reflexes.

Disease severity is highly variable, even within the same kinship. Some individuals may show minimal signs and are unaware of being affected, whereas others may be significantly disabled. The clinical variability and genetic heterogeneity in CMT often poses difficult diagnostic challenges. To date, Shy, M.E. (3), and Szigeti, K. et al. (4) discloses that about 50 genes have been identified for CMT diseases.

However, the clinical phenotype is similar for almost all CMT subtypes, caused by mutations in many different genes, involved in very diverse functions. In consequence, the different forms of CMT are usually clinically indistinguishable in a single patient, and the precise CMT subtype diagnosis requires a complex approach, based on inheritance pattern, clinical and electrophysiological examination, and molecular analyses.

Most of the molecular approaches currently used for genetic analysis in CMT correspond to gene by gene explorations, starting by the most prevalent one. They are carried out on DNA extracted from blood samples according to the following steps:

(1 ) a molecular analysis of the most prevalent gene 1 is implemented, in order to identify abnormality on said gene 1 ; and

(2) if no abnormality is detected, after a clinical consultation, a molecular analysis is implemented on the second most prevalent gene 2, in order to search a possible abnormality on said gene 2.

The above steps are repeated for prevalent genes in decreasing order of prevalence, until a mutation is detected, giving a genetic confirmation of disease, before turning to therapy phase.

The total time necessary for finding relevant mutation can be up to several years. Using this gene by gene approach, more than half of the patients remain without molecular diagnosis, either because the mutations they carry in already known genes, escape the diagnostic methods currently used, or, more likely, for most patients, because their disease causing mutations lie in not yet identified genes.

Thus with the presently available technologies, a differential molecular genotyping is required, which is highly complex and time consuming (some weeks up to several years).

Piluso et al. (5) disclose a comparative genomic hybridization microarray for copy number variations in 245 genes and 180 candidate genes implicated in NMDs, among which 26 are known CMT genes. Even though the method of Piluso et al. allows detecting copy number variations, said method does not allow detecting all the molecular causes of Inherited Neuromuscular Disorders, especially for CMT.

It would therefore be desirable to provide a method for determining all the molecular causes of Inherited Neuromuscular Disorders more rapidly, more accurately and at lower cost than the prior art techniques, particularly for CMT diseases

SUMMARY OF THE INVENTION

The applicant has now found that such an aim is achieved by an in vitro method of identifying molecular causes for CMT disease.

As used herein, "molecular causes" indicates mutations due to Copy Number Variations (hereafter CNVs) and point mutations.

More precisely, the present invention relates to a method of identifying in vitro molecular causes of CMT disease, comprising the following steps:

(i) providing a physiological sample comprising a genome of a subject, and

(ii) implementing on said sample process A or process B or both processes, wherein:

- Process A comprises determining a number of copy number variation(s), with respect to a sample of a normal subject, on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1 and

- Process B comprises determining a number of point mutation(s) with respect to a sample of a normal subject on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1 .

In order to detect at least one of the CMT subtypes, it is mandatory to screen for mutations in all different targeted genes.

Two types of gene mutations are involved in said CMT:

1 . CNV, and

2. Point mutation.

As used herein, "Copy Number Variation (CNV)", is an alteration of the DNA of a genome, resulting in an abnormal number of copies of one or more sections of the DNA. CNV can be a gain or a loss of specific DNA sequence(s) in DNA, such as deletions, duplications or amplifications of sequence(s).

As used herein, "point mutation" is replacement of single base nucleotide with another nucleotide of the genetic material.

As used herein, "a normal subject" indicates a subject who is devoid of any neuromuscular disease.

Process A allows determining a number of CNV(s) on at least one of the targeted genes of Table 1 , and thus determining a number of at least one of the CMT subtypes arising from CNV(s).

In one embodiment of a method of identifying in vitro molecular causes of Charcot-Marie-Tooth disease of the present invention, step (ii) comprises only Process A.

Process A however does not allow detecting (a) point mutation(s) on said genes.

Process B allows detecting (a) point mutation(s) on said genes of Table

1 , and thus determining a number of at least one of the CMT subtypes arising from (a) point mutation(s).

In another embodiment of a method of identifying in vitro molecular causes of CMT disease of the present invention, step (ii) comprises only Process B.

Therefore, it is interesting to implement Process B in order to detect (a) point mutation(s) on said genes, thus enabling to determine a number of at least one of the CMT subtypes with a higher rate of determination with respect to a well-known technique of the prior art and any conventional technique.

Particularly, since Process A and Process B are complementary, their combined use allows determining all possible molecular causes on said genes with a high determination rate.

In a preferred embodiment of a method of identifying in vitro molecular causes of CMT disease of the present invention, step (ii) comprises Process A and Process B.

Table 1 hereafter presents 53 genes which have been identified, after extensive research on CMT, to be involved in at least one of the CMT subtypes.

The definition of each gene can be found on http://genome.ucsc.edu. PMP22 PRPS1

MPZ HK1

GJB1 FIG4

MFN2 CTDP1

SIMPLE/ LITAF KIFlB

EGR2 LMNA

NEFL MED25

GDAP1 YARS

MTMR2 BSCL2

MTMR13/SBF2 DCTN1

SH3TC2 SETX

NDRG1 HSN2/WNK1

PRX IKBKAP

FGD4 NGFB

RAB7 IGHMBP2

GARS PLEKHG5

HSPB1 SEPT9

HSPB8 SLC12A6

DNM2 SOX10

GAN ARHGEF10

SPTLC1 HOXD10

NTRK1 AARS

TRPV4 LRSAM1

Group 1 FAM134B

KARS

SCN9A

CCT5

SPTLC2

ATL1

FBLN5

Group 2

Table 1

In Table 1 , 23 genes are classified as Group 1 corresponding to high level of implication, 30 genes are classified as Group 2 corresponding to a certain level of implication, in at least one of the CMT subtypes.

Table 2 shows the implication of the genes of Table 1 in each CMT suptybe. Chromosomal

Gene symbol Name CMT subtype Localisation

PMP22 peripheral myelin protein 22 CMT1A/H N PP 17pll.2

MPZ myelin protein zero CMT1B lq23.3

lipopolysaccharide-induced

SIMPLE/ LITAF TN F factor CMT1C 16pl3.13 early growth response 2

EGR2 protein CMT1D 10q21.2 neurofilament, light

NEFL polypeptide 68kDa CMT1F/CMT2E 8p21.2

GJB1 connexin 32 CMTX1 Xql3.1

phosphoribosyl pyrophosphate

PRPS1 synthetase 1 CMTX5 Xq22.3

ganglioside-induced

GDAP1 differentiation-associated CMT4A/AR-CMT2C 8q21.11

MTMR2 myotubularin-related protein 2 CMT4B1 llq21

MTMR13/SBF2 SET binding factor 2 CMT4B2 llpl5.4

SH3 domain and

KIAA1985/SH3TC2 tetratricopeptide repeats 2 CMT4C 5q33.1

N-myc downstream regulated

NDRG1 gene 1 CMT4D 8q24

PRX periaxin isoform 1 CMT4F 19ql3.2

CMT4G (H MSN-

HK1 Hexokinase 1 Russe) 10q22-q23

FYVE, RhoGEF and PH domain

FGD4 containing 4 CMT4H 12pll.21

Sac domain-containing inositol

FIG4 phosphatase 3 CMT4J 6q21

CTD (carboxy-terminal domain,

CTDP1 RNA polymerase I I, CCFDN 18q23

MFN2 mitofusine 2 CMT2A lp36.22

KIFlB Kinesin Family member IB CMT2A lp36.22

RAB7, member RAS oncogene

RAB7 family CMT2B 3q21.3

GARS glycyl-tRNA synthetase CMT2D 7pl5.1

HSPB1 heat shock 27kDa protein 1 CMT2F 7qll.23

HSPB8 heat shock 22kDa protein 8 CMT2L 12q24.23

LMNA lamine A/C AR-CMT2A/CMT2B1 lq22

MED25 mediator complex subunit 25, AR-CMT2B/CMT2B2 19ql3.33 subunit of the humanactivator- recruited cofactor (ARC)

DNM2 dynamin 2 DI-CMTB 19pl3.2

YARS tyrosyl-tRNA synthetase DI-CMTC lp35.1

BSCL2 seipin isoform 1 d-H M N-V llql2.3

DCTN1 dynactin 1 dH MN-VI IB 2pl3.1

dH MN+upper motor

SETX (ALS4) senataxin neuron (ALS4) 9q34.13 giant axonal

GAN gigaxonin neuropathy 16q23.2 serine palmitoyltransferase

SPTLC1 subunit 1 isoform a HSAN-I 9q22.31 hereditary sensory

HSN2/WNK1 neuropathy, type I I HSAN-I I 12pl3.33 inhibitor of kappa light

IKBKAP polypeptide gene HSAN-I II 9q31.3 neurotrophic tyrosine kinase,

35 NTRK1 receptor, type 1 HSAN-IV lq23.1

nerve growth factor, beta

36 NGFB polypeptide precursor HSAN-V lpl3.2

immunoglobulin mu binding DSMA1 ou SMARD1

37 IGHMBP2 protein 2 ou dH M N-VI llql3.2

pleckstrin homology domain

38 PLEKHG5 containing family G DSMA4 lp36.31

39 SEPT9 septin 9 H NA 17q25.2

solute carrier family 12, Andermann

40 SLC12A6 member 6 isoform c syndrome 15ql4

SRY (sex determining region Y)-

41 SOX10 box 10 PCWH 22ql3.1

Rho guanine nucleotide Slowed nerve

42 ARHGEF10 exchange factor 10 conduction velocities 8p23.3

CMT+congenital

43 HOXD10 Homeobox D10 vertical talus 2q31.1

Alanyl tRNA synthetase,

44 AARS cytoplasmic CMT2N 16q22

E3 ubiquitin-protein ligase

(Leucine-rich repeat and

sterile alpha motif containing 9q33.22-

45 LRSAM1 1) AR-CMT2D q34.11

Transient receptor potential

cation channel, subfamily 5,

46 TRPV4 member 4 CMT2C 12q23-q24

Famlily with sequence

47 FAM134B similarity 134, member B HSAN-I IB 5pl5.1

48 KARS Lysine tRNA synthetase RI-CMT1B 1 6q23.1

SODIU M CHAN N EL, VOLTAGE-

GATED, TYPE IX, ALPHA AR congenital

49 SCN9A (Navl.7) SUBU N IT; SCN9A indifference to pain 2q24.3

CHAPERON I N CONTAIN I NG

50 CCT5 TCP1, SU BUN IT 5 HSN+SPG 5pl5.2

serine palmitoyltransferase,

51 SPTLC2 long chain base subunit 2 HSAN5 14q24.3

52 ATL1 Atlastin 1 HSN 1+SPG3 14qll-q21

CMT1G+ age-related

macular

53 FBLN5 Fibulin 5 degeneration 14q32

Table 2: 53 genes of Table 1 classified by implications in each CMT subtype DESCRIPTION OF THE PREFERRED EMBODIMENTS

As shown in Figure 2, providing a blood sample is sufficient to implement the method of the present invention.

The present invention relates to a method of identifying in vitro molecular causes of CMT disease, comprising the following steps:

(i) providing a physiological sample comprising a genome of a subject, and

(ii) implementing on said sample at least one of Process A and Process B, wherein:

- Process A is determining a number of copy number variation(s), with respect to a sample of a normal subject, on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1 , and

- Process B is determining a number of point mutation(s), with respect to a sample of a normal subject, on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table

1 .

In one embodiment of a method of identifying in vitro molecular causes of CMT disease of the present invention, step (ii) comprises only Process A.

In another embodiment of a method of identifying in vitro molecular causes CMT disease of the present invention, step (ii) comprises only Process B.

In a preferred embodiment of a method of identifying in vitro molecular causes of CMT disease of the present invention, step (ii) comprises both Processes A and B.

As used herein, "targeted genes" signifies the genes on which Process

A or Process B is, or Process A and Process B are carried out.

Process A allows determining a number of CNV(s) on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1 .

Process B allows determining a number of point mutation(s) on all the

23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1

When the number of targeted genes increases, the determination rate of at least one of the CMT subtypes is increased. When step (ii) comprises both Process A and Process B, said at least 20 genes selected from the 30 genes of Group 2 in Table 1 for Process A can be selected dependently or independently from said at least 20 genes selected from the 30 genes of Group 2 in Table 1 for Process B. In other terms, the genes used in Process A and Process B may be the same or may be partly or totally different from each other.

In one preferred embodiment, Process A or Process B is, or Process A and Process B, are carried out on all the 53 genes in Table 1 .

In other preferred embodiments providing an excellent rate of detection, Process A or Process B is, or Process A and Process B are carried out on all the 49 genes of Table 3 hereunder:

Table 3 If one or more CNVs or one or more punctual mutations are(is) detected, the physiological sample comprising a genome of a subject is classified as positive and a precise CMT subtype can be allotted to said sample.

Process A consists in

- providing a physiological sample comprising a genome of a subject, and

- determining a number of CNV(s) on targeted genes of the genome of the sample.

From the CNV(s) detected on said targeted genes, a person skilled in the art can determine a CMT subtype arising from CNV(s) on said genes. For example, when a CNV is detected on gene PMP22, which is involved in both CMT1 A and HNPP, if the CNV is a duplication, said sample has CMT1 A arising from a CNV, and if the CNV is a deletion, said sample has HNPP arising from a CNV.

Here, Process A may be implemented by any well-known technique of the prior art and any conventional technique allowing determining a number of CNV(s).

In one embodiment, Process A is carried out with a Device A comprising a set of probes for said targeted genes.

In one particular embodiment, said Device A is a Comparative Genomic

Hybridization (CGH) array.

Therefore, for example, by using CGH technique, Process A consists in

- providing a physiological sample comprising a genome of a subject,

- providing a CGH array provided with a set of probes for all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the

30 genes of Group 2 in Table 1 ,

- analysing said sample using said CGH array, and

- determining a number of CNV(s) in the DNA of the sample.

A suitable physiological sample may be for example a biopsy sample, whole blood, a lymphocyte culture, preferably, whole blood or a lymphocyte culture, particularly preferably a lymphocyte culture.

One usual blood sampling provides an amount of sample sufficient for implementing the method of the present invention.

As used herein, "Comparative Genomic Hybridization (CGH)" is a type of nucleic acid hybridization assay which detects and identifies the location of CNV. The technique is described in detail, for example in the references (6) (7) and (8):

CGH is a co-hybridization assay of differentially labelled test DNA (for example green fluorescent dye) and reference DNA (for example red fluorescent dye) that includes the following major steps:

(1 ) immobilization of nucleic acids on a support to provide an immobilized probe;

(2) pre-hybridization treatment to increase accessibility of the probe and to reduce nonspecific binding; (3) hybridization of a mixture of target nucleic acids to the probe;

(4) post-hybridization washing to remove nucleic acid fragments not hybridized to the probe; and

(5) determination of the target nucleic acids hybridized to the probes using a determination device.

The 53 genes of Table 1 are involved in at least one of the CMT subtypes.

Device A comprises a set of probes for at least all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1 .

When the number of targeted genes increases, the determination rate of at least one of the CMT subtype is increased.

Therefore in one preferred embodiment, said Device A comprises a set of probes for all the 53 genes in Table 1 .

CNV(s) in a tested DNA of the sample can be determined in the following manner, using two colour labels:

In case of a deletion in the tested DNA, less tested DNA will bind to the corresponding spots and a first colour label of the reference DNA will prevail. Gains in the tested genome can be identified by a dominance of second colour label of the tested DNA. Spots representing sequences with the same copy number in the tested genome relative to the reference genome are revealed by a third colour.

Thus, the ratio fluorescence intensity of the tested DNA / fluorescence intensity of the reference DNA is then calculated, in order to measure the copy number changes for a particular location in the genome.

As used herein, "a set of probes" means a set of fragments of nucleotides having sequences capable of hybridizing with the sequence of the genes to be analysed.

A person skilled in the art can prepare a suitable set of probes for said Device A, by any well-known technique of the state of the art, such as the method described by Technical note (9).

In one embodiment, said set of probes for said Device A comprises: - probes evenly spaced by about 30 bp distance between the beginning of two consecutive probes, which hybridize the coding region (exons) of said gene plus 200 bp at each side of the exons,

- probes evenly spaced by about 1 0 bp distance between the beginning of two consecutive probes, which hybridize the 5'- and 3'-untranslated regions of said gene (3'-UTR and 5'-UTR),

- probes evenly spaced by about 60 bp distance between the beginning of two consecutive probes (adjacent probes) in a region of about 2000 bp at the 5' and 3' terminal exons and in introns,

- backbone probes not evenly spaced by about 30 000 bp distance between two consecutive probes, which represent on average 60% of the total probes on said device.

As used herein, "evenly spaced" indicates "spaced by a same interval". As used herein, "backbone probes" are probes which hybridize to locations on the genome going beyond the genes of interest, such as potentially intergenic regions. They are used to generate a calibration signal against which the test and reference signals from the specific gene probes are measured.

As used herein, "covered" indicates "capable of hybridizing with the sequence of the genes to be analysed".

In one particularly preferable embodiment, said set of probes are manufactured according to the following rules:

- genomic coordinates of probes refer to the hg1 8 genome assembly (1 0)

- probes cover gene +/- 2 kb up and downstream,

- probes are alternated on (+) and (-) strands,

- with a tiling of

- 30 pb tiling in exonic regions and intron-exon boundaries (200 bp upstream and downstream of the exon)

- 50 pb tiling in 3' and 5' UTR

- adjacent probes in introns and in 2000 bp regions upstream and downstream of said gene

- backbone probes each 30 kb,

- total number of probes: 1 35947.

- gene probes (exon, intron, 5' and 3'-UTR, regions +1-2 kb from gene):

51 553. - backbone probes: 84394 (one each 30 kb on average).

As used herein, "average probe density" indicates the inverse of the mean distances between the start positions of consecutive probe sequences on the indicated region of the genome:

n— 1

wherein n is the number of probes in the considered region of the genome.

As used herein, "tiling" indicates the mean distance between the beginning of two consecutive probes. This probe design allows increasing the robustness of the determination of the present invention.

Process B consists in

- providing a physiological sample comprising a genome of a subject, and

- determining a number of point gene mutation(s) on targeted genes of the genomes of the sample.

From said point mutation(s) detected on said targeted genes, a person skilled in the art can determine a number of at least one of the CMT subtype arising from (a) point mutation(s) on said genes.

For example, if a point mutation is detected on gene FGD4, which is involved in CMT4H subtype (c.f. Table 2), said sample has CMT4H arising from a point mutation.

Here, Process B may be implemented by any well-known technique of the prior art and any conventional technique allowing determining a number of point mutation(s).

In one embodiment, Process B is carried out with a Device B comprising a set of probes for said genes.

In one embodiment, Process B is carried out by a technique selected from the group consisting of Sequence capture, "on-chip capture" and "in- solution capture (Sure Select)".

In one particular embodiment, Device B is a Sequence capture array. Therefore, for example, Process B consists in

- providing a physiological sample comprising a genome of a subject,

- providing a sequence capture array provided with a set of probes for the exome enriched in probes for 53 genes listed in Table 1 , - analysing said sample using said sequence capture array, and

- determining a number of punctual mutation(s) in the DNA of the sample.

For Process B, the same physiological sample as that prepared for Process A may be used.

The 53 genes of Table 1 are involved in at least one of the CMT subtypes.

Said Device B comprises a set of probes for all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1 . When the number of targeted genes increases, the determination rate of a number of at least one of the CMT subtypes is increased.

In one more preferred embodiment, said Device B comprises a set of probes for all the 53 genes in Table 1 .

"DNA sequence capture" consists in isolating and sequencing a genomic region of interest (targeted region), to the exclusion of the remainder of the genome, and then sequencing the captured DNA fragments.

"Sequencing" means determining the sequence of target DNA fragments.

DNA sequence capture includes the following major steps:

(1 ) preparation of a sequencing library of test DNA;

(2) hybridization of said sequencing library to "in solution" probes thereon;

(3) post-hybridization washing to remove nucleic acid fragments not hybridized to the probe;

(4) target DNA fragment elution;

(5) PCR amplification of target DNA fragment; and

(6) sequencing of target DNA fragment.

In the present invention, the term "target regions" indicates regions of a gene which are especially involved in at least one of the CMT subtypes.

A person skilled in the art can manufacture a suitable set of probes for said Device B by any well-known technique of the prior art and any conventional technique, such as the method described by reference (6).

In one embodiment, said set of probes for said Device B comprises: - probes of 70 to 120 bp, which hybridize all the exons of said genes with at least 2X tiling frequency.

As used herein, "tiling frequency" indicates the density of tiling. For example, 2X tiling frequency means that each base is covered by two different probes.

In one embodiment, the set of probes for said Device B is "enriched" in the exome.

"Exome" means the subset of a genome that is protein coding.

"Enrich" means that the density of probes in the "enriched" region is higher than in the "non enriched" region. In other terms, it corresponds to adding more probes for at particular zones.

Enriching exome of probes allows a better capture of the corresponding genomic region, and as a consequence, better coverage a better depth of coverage at sequencing, as well as less dispersion of data.

Figure 3-A shows gene coverage in enriched (dark) versus non enriched

(bright) genes, and Figure 3-B shows depth of coverage in enriched (dark) versus non enriched genes.

In one embodiment, said set of probes for said Device B for Process B is prepared according to the following rules:

- 44.1 Mb are covered by probes from a commercial Exome library. We used the SeqCap EZ Human Exome Library v2.0 from Nimblegen (http://www.nimblegen.com/products/seqcap/ez/v2/index.html). The SeqCap EZ Exome Library v2.0 product covers more than 20,000 genes in the human genome. The following sources provided information about the genes:

• NCBI Reference Sequence (RefSeq) RefGene from UCSC (January 2010)

• CCDS from NCBI (September 2009)

• miRNAs from miRBase (version 14, September 2009)

· Customer inputs

All the SeqCap EZ Human Exome Library v2.0 Choice XL genome coordinates were based on human genome build GRCh37 (hg19). For RefSeq genes, only transcripts with an "NM_" prefix were selected, and only protein coding parts of the transcripts were targeted. For exons that are smaller than 100 bp, Roche NimbleGen extended the target region to 100 bp.

The total size of the target regions is 36.5 Mb. Roche NimbleGen selected 2.1 million long oligo probes to cover the target regions. Because some flanking regions are also covered by probes, the total size of regions covered by probes is 44.1 Mb, larger than the initial target regions. Sequences of the probes can be found at http://www.nimblegen.com/products/seqcap/ez/v2/index.html.

-Additional custom probes targeting 53 CMT genes from the list provided in Table 2 (all genes except FBLN5) have been added to the SeqCap EZ Human Exome Library v2.0 Choice XL from Nimblegen. The probes target a total of 1 .75 Mb of DNA. They have been designed according to the following rules

- All exonic regions are covered including 5'UTRs, coding exons and

3'UTRs,

- Introns, upstream and downstream sequences from genes are not directly covered by probes.

This probe design allows increasing the robustness of the determination of the present invention.

Device B of the present invention preferably comprises an array or solid particles suspended in liquid such as magnetic particles.

In one embodiment, High Throughput Sequencing (hereafter HTS) is used in Process B, for allowing further lowering the cost of analysis.

Device A may be manufactured as follows:

The sequences of targeted genes known to be responsible for at least one of the CMT subtypes were obtained from the web site of the UCSC (http://genome.ucsc.edu) and are shown in Table 1 .

The genes on which analysis is implemented are selected from Table 1 , as explained above.

A set of probes for Device A can be designed by a well known technique of the state of the art, as explained above. For example, CGH arrays which may be used in Process A may be manufactured by any manufacturer specialized in preparation of such arrays, such as Roche-Nimblegen.

After completion of the preparation of a set of probes, the probes may be fixed on the support to prepare a CGH array.

In one embodiment, a CGH array containing a set of probes for all the 53 genes in Table 1 was prepared.

This approach is consistent with the chosen technology, allowing the spot of up to 12 x 135 000 probes in the same array, thus identifying at a glance deletions or duplications in all the known genes with a high resolution (one probe every 10 bases or so).

Device B may be manufactured as follows:

Sequences of targeted genes known to be responsible for at least one of the CMT subtypes were obtained from the web site of the UCSC and are shown in Table 1 .

A set of probes for Device B is designed by a well known technique of the state of the art, as explained above.

For example, sequence capture arrays which may be used in Process B may be manufactured by any manufacturer specialized in preparation of such arrays, such as Roche-Nimblegen or Agilent.

In one embodiment, the set of probes is enriched in the exome.

After completion of the preparation of a set of probes, the probes may be fixed on the support to prepare a sequence capture array or used "in solution".

In one preferred embodiment, a sequence capture array containing a set of probes for the Human exome (more than 20 000 genes) with specific enrichment of probes for 53 genes in Table 2 was prepared.

This approach is consistent with the chosen technology, allowing, theoretically, the capture, in solution, of all the coding sequences from the Human Genome, with a better efficacy of capture at regions corresponding to 53 CMT genes (listed in Table 2). The sequences captured by this array, are then desorbed and characterized by HTS. Advantage of HTS is that, for example, 2-3 patients could be pooled and sequenced simultaneously decreasing by 2-3 the cost of mutation identification.

The methods according to the invention have advantageous properties.

Since Process A is carried out on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1 , Process A allows a CMT determination rate of at least 30%.

When Process A is carried out on all the 53 genes in Table 1 , Process A allows a higher CMT determination rate.

Since Process B is carried out on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1 , Process B allows a CMT determination rate of at least 30%.

When Process B is carried out on all the 53 genes in Table 1 , Process B, allows a higher CMT determination rate

As used herein, the determination rate is the percentage of patients for whom the causative mutation(s) could be identified in the group of tested patients.

A combined determination of Process A and Process B allows detecting both a number of CNV(s) and a number of point mutation(s) on the targeted genes, and therefore determining a number of at least one of the CMT subtypes.

In particular, said combined determination allows increasing the rate of determination of a number of at least one of the CMT subtypes.

The present challenge lies in increasing determination rate via characterisation of all mutation types, allowing characterization of the genotype also in rare and atypical phenotypes, in genetically ambiguous sporadic cases and in CMTs whose pathophysiology is multiallelic or multigenic.

By a gene by gene approach analysis of the art of the technique, about 50% of patients suffering from CMT are devoid of genetic confirmation. In contrast, by the analytic method of the present invention, the rate of non- determined patients, i.e. allotted patients for whom no causative mutation could be identified, is considerably less than 50% when Process A and Process B are carried out on 53 genes. Additionally, the time necessary to obtain complete results is dramatically reduced from several years to a couple of weeks. Because of the special selection of the group of genes of Table 1 , involved in at least one of the CMT subtypes, and of the specific Devices A and B both provided with a set of probes for said group of genes, the process of the present invention allows determining all possible mutations in said genes and therefore determining a number of at least one of the CMT subtypes, with a rate of determination of at least 90%, often at least 95%, and generally at least 99%, when Process A and Process B are carried out on all the 53 genes of Table 1 .

A further subject matter of the present invention relates to a method for determining at least one of the CMT subtypes, comprising implementing the above mentioned step(s) of determination.

The method of the present invention can be applied to the determination of a number of CNV(s) or point mutation(s), or CNV(s) and point mutation(s) arising from at least one of the CMT subtypes.

In one embodiment of a method of identifying in vitro molecular causes of Inherited Neuromuscular Disorders of the present invention, that is to say CMT, step (ii) comprises only Process A.

In another embodiment of a method of identifying in vitro molecular causes of Inherited Neuromuscular Disorders of the present invention, that is to say CMT, step (ii) comprises only Process B.

In a preferred embodiment of a method of identifying in vitro molecular causes of Inherited Neuromuscular Disorders of the present invention, step (ii) comprises both Process A and Process B.

If one or more CNVs or one or more punctual mutations are(is) detected, the physiological sample comprising a genome of a subject is classified as positive and a precise type of at least one of the CMT subtypes is allotted to said sample.

The present invention provides sensitive and reliable tools for detecting at least one of the CMT subtypes with a high determination rate, as evidenced by the Examples.

Said determination rate is at least 90%, often at least 95%, and generally at least 99%.

They allow determining all possible mutation(s) involved in at least one of the CMT subtypes, in targeted gene(s) in only a single step process (2,100,000 probes). This is possible due to the specificity of the selection of the group of genes of Table 1 , involved in at least one of the CMT subtypes, and the specificity of Device A and Device B, both provided with a set of probes for said groups of genes, with a high determination rate. Said determination rate is at least 90%, often at least 95%, and generally at least 99%, when Process A and Process B are carried out on all the 53 genes of Table 1 .

The process of the present invention increases the ratio of precisely analysed patients (either new analysed patients or reoriented patients for which initial analysis was erroneous).

Therefore the process of the present invention allows improving genetic counselling and patient management, establishing phenotype-genotype correlations, constructing dedicated databases and including patients in current or future clinical trials due to the special selection of groups of genes to be analysed.

The present invention allows also reducing analysis costs by the special selection of group of genes to be analysed, and by using platforms with high analytic capacities. The method of the present invention really corresponds to a "one-shot" technology that considerably reduces both the time and the cost of the whole analytic process.

Such an approach also dramatically reduces the costs associated to analysis on the medium term. The special selection of groups of genes to be analysed allows reducing the time-to-analysis (down to 72h to one week) for patients and families.

One must keep in mind that patients with a well characterized pathology, both on clinical and genetic sides, are the only one eligible for clinical trials or protocols, which become more and more numerous.

The scope of the invention can be understood better by referring to the examples given below, the aim of which is to explain the advantages of the invention. The invention will now be described by means of the following Examples. DESCRIPTION OF THE DRAWINGS

Figure 1 shows a flow chart depicting the major steps involved in detecting at least one of the CMT subtypes, using the gene by gene exploration of the prior art.

Figure 2 shows a flow chart depicting the major steps involved in detecting at least one of the CMT subtypes using a method of the present invention.

Figure 3-A shows gene coverage in enriched (dark) versus non enriched (bright) genes, and Figure 3-B shows depth of coverage in enriched (dark) versus non enriched genes.

Figure 4 shows the result of analysis of two samples from one patient affected with CMT1 A (Sample A) and one patient affected with HNPP (Sample B), using a CGH array. These patients present the "classical" duplication/deletion of PMP22 region.

Figure 5 shows the result of analysis of samples from patients affected with CMT1 A (Samples A and B) or HNPP (Samples C and D), using a CGH array. These patients present the "classical" duplication/deletion of PMP22 region or rarer smaller partial deletions in the same region.

Figures 6 shows the result of analysis of one sample from a patient affected with CMT4D (HMSN-Lom).

Figure 7 shows the results of identification of a genomic variation in a patient using sequence capture and lllumina sequencing.

EXAMPLES EXAMPLE 1 : MANUFACTURE OF DEVICES

A CGH array and a sequence capture array ware prepared by Digital Mirror Device according to methods known in the art. The Digital Mirror Device creates "virtual masks" that replace physical chromium masks used in traditional arrays.

These "virtual masks" reflect the desired pattern of UV light with individually addressable aluminium mirrors controlled by the computer. The DMD controls the pattern of UV light projected on the microscope slide in the reaction chamber, which is coupled to the DNA synthesizer. The UV light selectively cleaves a UV-labile protecting group at the precise location where the next nucleotide will be coupled. The patterns are coordinated with the DNA synthesis chemistry in a parallel, combinatorial manner such that 385,000 to 4.2 million unique probe features are synthesized in a single array.

The set of probes for CGH array has been selected according to the following rules:

- genomic coordinates of probes refer to the hg1 8 genome assembly

- probes cover gene +/- 2 kb up and downstream,

- probes are alternated on (+) and (-) strands,

- with a tiling of

- 50 pb tiling in 3' and 5' UTR

- backbone probes each 30 kb,

- total number of probes: 1 35947.

- gene probes (exon, intron, 5' and 3'-UTR, regions +1-2 kb from gene):

51 553.

- backbone probes: 84394 (one each 30 kb on average).

The set of probes for sequence capture array has been selected according to the following rules:

- All exonic regions are covered including 5'UTRs, coding exons and 3'UTRs,

EXAMPLE 2: CGH array NimbleGen

Starting Material :

Genomic DNA from patients is used as starting material for CGH analysis. It will be compared to a reference DNA corresponding to a pool of anonymous donors (Promega, G1 521 ).

The extraction protocol recommended for DNA purification is the Qiagen DNeasy Blood & Tissue Kit. (Qiagen, 50x, cat. no. 69504). Optional RNase treatment step must be achieved for these applications (RNase A, 100 mg/ml, cat. no. 19101 ).

Quality Control:

DNA are assessed for quality and concentration using respectively agarose gel and spectrophotometer method (Nanodrop, ND-1000). RNase A treatment is recommended as RNA contamination could interfere during hybridization.

The quality control of genomic DNA is based on Nanodrop spectrophotometer measurements. Absorbance at 260nm (A260) is used to assess quantity and A260/A280, A260/A230 ratios are calculated to assess purity of samples.

These ratios should be as follows: A260/A230 > 1 .9

A260/A280 > 1 .8

To determine integrity of DNA, 250 ng should be analyzed on a 1 % agarose gel to ensure that they show no sign of RNA contamination or degradation.

DNA labelling:

Each genomic DNA from patients is labelled using known fragment allowing incorporating cyanin 3 (Cy3) in the DNA. In the same time, reference DNA is labelled using the same method except that cyanin 5 (Cy5) is incorporated instead of Cy3. Experiments are implemented in "Dye swap": in the dye swap experiment the DNA from the patient is labelled with cyanin 3 and the reference DNA labelled with cyanin 5.

Hybridization:

Same quantity of labelled DNA is then pooled (1 DNA from patient with the reference DNA) in a suitable hybridization cocktail and added into a hermetic chamber on the array. Hybridization is led during 3 days at 42 °C using the NimbleGen hybridization Station.

Washing and scanning:

Arrays are washed using several buffers by increasing stringency and finally dried. Slides are then scanned twice in a 2 μιτι-resolution scanner (MS200 NimbleGen) using 2 wave-lines corresponding to each fluorochrome. Data analysis:

Quality control check and grid alignment are performed on scanned images. NimbleScan software is then used to convert intensity into raw data files and calculate log2 (ratio) corresponding DNA from patients normalized by the reference. Ratio data are in .gff format file allowing visualizing the results using a genome browser (Signal Map) or other third part tools (like CGH-web).

For a detailed protocol see:

http://www.nimblegen.com/products/lit/NG_CGHCNV_Guide_v8pO.pdf 2-1 : Duplication/deletion of PMP22 region in CMT1A (Process A): "classical" duplication/deletion (Process A)

Blood samples were taken from patients suffering of (i) autosomal dominant demyelinating CMT (CMT1 ) (sample A), and (ii) HNPP (Hereditary Neuropathy with liability to Pressure Palsies (sample B) . Both samples were analysed, in a dye-swap experiment, using a CGH array provided with a set of probes for the 53 genes of Table 1 .

As shown in the signal map visualization of Figure 4, a heterozygous duplication of the whole PMP22 gene and of a 1 .5 Mb surrounding region was detected in sample A: DNA from patient affected with CMT1 A. The "mirror" heterozygous deletion (genomic size 1 .5 Mb) was detected in sample B: DNA from patient suffering of HNPP.

In Figure 4, the abscissa indicates genomic coordinates, and the ordinate indicates the log2 ratio of signals.

The breakpoints determined by CGH are Chr17: 31827185-32000205.

2-2: Partial duplications of the P/WP22CMT1 A region in CMT1 A patients (Process A)

Blood samples were taken from patients suffering of autosomal dominant demyelinating CMT (CMT1 ) (sample A and B), and from patients affected with HNPP (samples C and D) (see Figure 5). These samples were analysed with a CGH array provided with a set of probes for the 53 genes of Table 1 .

In samples A and D, "classical duplications/deletions" (see point 2.1 ) have been detected. In samples B and C, duplications and deletions, respectively, smaller than the "classical" 1 .5 MB duplication/deletion have been detected (Figure 5):

• A 382-619 kb duplication removing the entire PMP22 gene in sample A (distal breakpoint chrl 7:14949365 and Chr17:14981 150 and proximal breakpoint comprised between Chr17:15533489 and Chr17: 15362369).

• A 7.7-38.5 kb deletion removing exons 1 -3 of PMP22 \n sample C (distal breakpoint chrl 7:15104010 and proximal breakpoint comprised between Chr17:151 1 1753 and Chrl 7: 15142556). 2-3: Deletion in NDRG1 (Process A)

A blood sample was taken from one female patient suffering of demyelinating CMT, autosomal recessive (CMT4). This sample was analysed with a CGH array provided with a set of probes for the 53 genes of Table 1 .

We have detected a homozygous deletion encompassing exons 9 to 13 in the NDRG1 gene (responsible for CMT4G/HMSN-Lom) (Figure 6).

In Figure 6, the abscissa indicates genomic coordinates, and the ordinate indicates the log2 ratio of signals. The upper panel shows the result of Upper panel: Cyanine 5 labeling of DNA from patient showing the deletion with a log2 ratio of -1 .3. lower panel shows the result of the dye swap experiment (DNA from patient labeled with Cy3), with mirror duplication with a log2 ratio of +1 .6

The breakpoints determined by CGH are Chr8:134,326,855- 134,337,524. Breakpoints have been confirmed by PCR. The deletion results in an in-frame deletion of 106 AA in the protein (aa 180 to 286): c.537+549_856- 1044del (p.Lys179_Met286del106). The deletion is likely pathogenic and causing CMT4G or HMSN-Lom in this patient.

EXEMPLE 3: Sequence Capture array (Process B)

In order to focus the sequencing on regions of interest, sequence capture strategies have been developed. In this aim, Agilent and Nimblegen propose in-solution capture methods that use specific oligonucleotide probes targeting regions of interest. These probes libraries are used to selectively capture DNA fragments from patients corresponding to these regions addressed by the sequencing. Here, next-generation sequencer HiSeq2000 from lllumina is used.

Agilent protocol:

(SureSelect Target Enrichment Kit, Agilent)

Quality Control:

DNAs are assessed for quality and concentration using respectively agarose gel and spectrophotometer method (Nanodrop, ND-1000). RNase A treatment is recommended as RNA contamination could interfere during hybridization.

The quality control of genomic DNA is based on Nanodrop spectrophotometer measurements. Absorbance at 260nm (A260) was used to assess quantity and A260/A280, A260/A230 ratios were calculated to assess purity of samples.

These ratios should be as follows: A260/A230 > 1 .9

A260/A280 > 1 .8

To determine integrity of DNA, 250 ng was analyzed on a 1 % agarose gel to ensure that they show no sign of RNA contamination or degradation.

DNA preparation:

Genomic DNAs from patients were fragmented using Covaris device to obtain fragments between 150 and 200bp length. Then, DNAs were repaired in order to get blunt-ended fragments that will be subjected to an addition of 5'- Phosphate before a 3'-end adenylation step. These modifications allow the ligation of specific adaptors (TruSeq adaptors, lllumina) that are used during the sequencing step. To obtain enough material for the hybridization step, a PCR amplification using Herculase II Fusion DNA Polymerase (Ozyme) was performed.

Hybridization on in-solution library probes:

DNA fragments were then hybridized 24h at 65 °C in presence of the capture probes. Purification of captured DNA:

Magnetic beads selection allows discarding of unspecific DNA and eluting the DNA fragments of interest. At the end of this step, a PCR amplification step was performed in order to amplify the material and to incorporate the specific TruSeq index sequences (lllumina). This barcode system allows pooling of DNA from different patients and sequencing of all of them in a single sequencing run. Quantity Assessment:

Quantification was performed using a qPCR kit (NGS Library Quantification, Agilent) in order to pool the samples in equimolar quantity. The pool was then ready for sequencing on HiSeq2000 platform.

For a detailed protocol see:

http://www.chem.agilent.com/Library/usermanuals/Public/G7530- 90000_SureSelectJlluminaXTMultiplexed_v.1 .2.pdf

NimbleGen protocol: (SeqEZ library, NimbleGen)

Protocol is very similar to the one use for Agilent SureSelect except for the following few points:

- The size of DNA fragments after fragmentation must be between 200 and 400bp length.

- The TruSeq indexes are added by ligation, before hybridization.

- Amplification is done using Fusion HF PCR master mix (Ozyme).

- Hybridization is performed at 47°C for 72h.

- Enrichment is assessed using qPCR methods in order to validate capture efficiency.

For a detailed protocol see:

http://www.nimblegen.com/products/lit/06588786001_SeqCapEZLibrarySR_Gui de_v3p0.pdf

3-1. Identification of a genomic variation (Process B)

Figure 7 shows the result of identification of a genomic variation in a patient using a custom "enriched" NimbleGen SeqCap EZ Human Exome v2.0 sequence capture and lllumina sequencing. Several reads from the sequencing of one patient have been aligned on the reference genome sequence (hg19 human genome). A heterozygous T>C variation is shown. Sequence reads generated by NGS sequencing are represented by horizontal grey bars. The reference sequence is displayed at the bottom as well as the corresponding amino acids.

< References >

(1 ) Skre, H. Genetic and clinical aspects of Charcot-Marie-Tooth's disease. Clin Genet 6, 98-1 18 (1974).

(2) Pareyson, D., Scaioli, V. & Laura, M. Clinical and electrophysiological aspects of Charcot-Marie-Tooth disease. Neuromolecular Med 8, 3-22 (2006).

(3) Shy, M.E. Charcot-Marie-Tooth disease: an update. Curr Opin Neurol 17, 579-85 (2004).

(4) Szigeti, K. & Lupski, J.R. Charcot-Marie-Tooth disease. Eur J Hum Genet 17, 703-10 (2009).

(5) Piluso G, Dionisi M, Del Vacchio Blanco F, Torella A, Aurino S, Savarese M, Giugliano T, Bertini E, Terracciano A, Vainzof M, Criscuolo C, Politano L, Casali C, Santorelli FM, and Nigra V. Motor chip: a comparative genomic hybridization microarray for copy-number mutations in 245 neuromuscular disorders. Clin. Chem. 201 1 : 57:1 1 1584-1596.

(6) Saillour Y, Cossee M, Leturcq F, Vasson A, Beugnet C, Poirier K, Commere V, Sublemontier S, Viel M, Letourneur F, Barbot JC, Deburgrave N, Chelly J, and Bienvenu T. Detection of exonic copy-number changes using a highly efficient oligonucleotide-based comparative genomic hybridization-array method. Hum. Mutat. 2008 Sep;29(9):1083-90.

(7) Hegde MR, Chin EL, Mulle JG, Okou DT, Warren ST, Zwick ME. Microarray- based mutation detection in the dystrophin gene. Hum. Mutat. 2008 Sep;29(9):1091 -9.

(8) Shinawi M, Cheung SW. The array CGH and its clinical applications. Drug Discov Today. 2008 Sep;13(17-18):760-70. Epub 2008 Jul 17. Review.

(9) Technical note : Roche NimbleGen Probe Design Fundamentals. 2008 June (http://www.nimblegen.com/index.html

(10) http://genome.ucsc.edu/cgi- bin/hgGateway?hgsid=326350413&clade=mammal&org=Human&db=hg18

Claims

1 . A method of identifying in vitro molecular causes of Charcot Marie Tooth disease, comprising the following steps:

(i) providing a physiological sample comprising a genome of a subject, and (ii) implementing on said sample process A or process B or both processes, wherein

- Process A comprises determining a number of copy number variation(s), with respect to a sample of a normal subject, on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of

Group 2 in Table 1 ; and

- Process B comprises determining a number of point mutation(s), with respect to a sample of a normal subject, on all the 23 genes of Group 1 in Table 1 and at least 20 genes selected from the 30 genes of Group 2 in Table 1 .

2. The method according to claim 1 , wherein said step (ii) comprises only Process A.

3. The method according to claim 1 , wherein said step (ii) comprises only Process B.

4. The method according to claim 1 , wherein said step (ii) comprises Process A and Process B.

5. The method according to any one of the preceding claims, wherein Process A or Process B is, or Process A and Process B are carried out on all the 53 genes in Table 1 .

6. The method according to any one of claims 1 , 2, 4 and 5, wherein Process A is carried out with a Device A comprising a set of probes for said genes.

7. The method according to claim 6, wherein said set of probes for said Device A comprises: - probes evenly spaced by about 50 bp distance between two consecutive probes, which hybridize said gene plus a region of about 2000 bp at the 5' and 3' terminal exons, and

- backbone probes not evenly spaced by about 30 kb distance between two consecutive probes, which represent on average of half of the total probes on said device.

8. The method according to claim 6 or 7, wherein said set of probes are manufactured according to the following rules:

- probes cover gene +/- 2 kb up and downstream

- probes are alternated on (+) and (-) strands

- with a tiling of

- 50 pb tiling in 3' and 5' UTR and 2kb regions upstream and downstream from the gene

-adjacent ad probes for in introns

- backbone probes each 30 kb

- total number of probes: 137207

- gene probes (exon, intron, 5' and 3'-UTR): 69570

- backbone probes: 67637 probes (one each 30 kb on average).

9. The method according to any one of the preceding claims, wherein Process A or Process B is, or Process A and Process B are carried out on all the 49 genes of Table 3:

SH3TC2 TRPV4 HINT1 SPTLC2

NDRG1 PRPS1 IKBKAP ATL 1

Table 3

10. The method according to any one of claims 1 , 2 and 4-9, wherein said Device A for Process A is a Comparative Genome Hybridization array.

1 1 . The method according to any one of claims 1 and 3-9, wherein Process B is carried out with a Device B comprising a set of probes for said genes.

12. The method according to claim 1 1 , wherein said set of probes for Device B is enriched in exome.

13. The method according to claim 1 1 or 12, wherein said set of probes for said Device B comprises:

- probes of 70 to 120 pb, which hybridize all the exons of said genes with at least 2X tiling frequency.

14. The method according to any one of claims claim 1 1 -13, wherein said set of probes for said Device B for Process B is prepared according to the following rules:

- All exonic regions are covered including 5'UTRs, coding exons and 3'UTRs,

15. The method according to any one of claims 1 and 3-14, characterized in that Process B is carried out by a technique selected from the group consisting of sequence capture, on-chip capture and in-solution capture.

16. The method according to any one of claims 15, characterized in that said Device B is a Sequence capture array.

17. The method according to any one of claims 1 , and 3-15 characterized in that High Throughput Sequencing is used in Process B.