AU2022218581A1

AU2022218581A1 - Sequencing data-based itd mutation ratio detecting apparatus and method

Info

Publication number: AU2022218581A1
Application number: AU2022218581A
Authority: AU
Inventors: Ruilin JING; Dawei Li; Hailiang Wang; Juan Wang; Zhaoling Xuan
Original assignee: ANNOROAD GENE TECHNOLOGY BEIJING CO Ltd
Current assignee: Annoroad Gene Technology (beijing) Co Ltd
Priority date: 2017-12-21
Filing date: 2022-08-18
Publication date: 2022-09-15
Anticipated expiration: 2038-12-20
Also published as: NZ766350A; WO2019120254A1; AU2018391843A1; AU2018391843B2; AU2022218581B2; CN109943635A

Abstract

A method for the quantitative ITD mutation detection based on sequencing data. The method comprises: acquiring sequencing data of samples to be detected; extracting ITD characteristics of sample to be detected; and obtaining the quantitative detection result on the basis of the ITD characteristic coefficients and ITD characteristics of samples to be detected.

Description

SEQUENCING DATA-BASED ITD MUTATION RATIO DETECTING APPARATUS AND

METHOD The present application is a divisional application from Australian Patent Application No.

2018391843, the entire disclosure of which is incorporated herein by reference.

TECHNICAL FIELD

The present invention relates to the field of gene mutation detection, and particularly relates to

a method, an apparatus and an electronic device for the quantitative detection of ITD mutation

based on sequencing data.

BACKGROUND

With the rapid development of next-generation sequencing technology and its increasing

involvement in scientific research and clinical detection of the field of cancer; we have reached a

new level of understanding of their occurrence and development, clinical manifestations and

pathogenesis. Numerous studies have shown that the occurrence of cancer is closely related to

somatic mutations, and these mutations often appear solely in subclones of certain tumors. The

study of subclonal mutations provides a new direction for disease progression and prognostic

stratification.

High-depth next-generation sequencing technology can detect subclonal mutations. By

sequencing specific genome regions using the target sequence capture sequencing technology, great

depth with low cost can be achieved, thereby a large number of sample data can be accumulated,

and this provides a favorable condition for accurate estimation of the distribution of false positive

rate at specific mutation sites in the genome.

The most common type of FLT3 gene mutation in acute myeloid leukemia (AML) is the

internal tandem duplication (ITD), followed by the Tyrosine-kinase domain (TKD) mutation.

FLT3-ITD mutation usually involves exon 14 and 15, or exon 11 and/or 12. AML patients are often

found to have internal tandem duplications (ITDs) in the juxtamembrane (JM) region of FLT3, i.e.

insertion of several repeating elements of oligonucleotide in an end-to-end order, these elements can

be copies of uncertain number of nucleotide bases but the number of bases are usually a multiple of

3 so that the reading frame remains intact, thereby extending the JM region, while other domains are unaffected. These mutations play an important role in the pathogenesis of AML. The incidence rate of FLT3-ITD in AML patients is about 24% for adults, 10%-15% for children, and about 15% for secondary AML patients. The NCCN Clinical Practice Guidelines in Oncology: Acute Myeloid Leukemia (2016) pointed out that the prognosis of patients carrying FLT3-ITD mutations with normal karyotype is poor and should be stratified into the high-risk group. Also, in the section about the treatment of patients with relapsed refractory AML, it was mentioned that patients with FLT3-ITD mutation may consider using a demethylating drug (5-azacytidine or decitabine) together with sorafenib [NCCN-AML]. Therefore, quantitative detection of FLT3-ITD is essential for AML patients. Currently, the predominant FLT3-ITD detection method is the cDNA-based PCR amplification. Limitation of this approach is that only quantitative detection can be performed, the position and sequence information of the ITD mutation, however, cannot be acquired simultaneously. If the acquisition of sequence information is needed, sanger sequencing must be performed and the result can be acquired only when the allele frequency of the ITD mutation is above 10%. That is, for patients with ITD mutation ratio below 10%, only the quantitative information can be obtained and the sequence and position information cannot be gained. Another FLT3-ITD detection method is to use NGS sequencing and ITD mutation is determined by an bioinformatics algorithm (e.g., PINDEL). This method is based on the high-depth target sequence capture sequencing, and it is realized by using the information from captured target regions. However, since the capture process may lose the target region in where the ITD is located, any presently available NGS-based sequencing algorithm (Pindel et al.) is therefore only suitable for qualitative analysis, and accurate quantitative results cannot be obtained. In addition, this method also has the following inevitable inherent limitations: 1. due to the template sequence content of ITD is significantly different from the normal genome, it has a great impact on the capture process. It is possible that relatively high proportion of ITDs could end up with only limited sequencing reads to support the existence of the mutation after target capture, and so it is hard to reach a correct conclusion; 2. also due to the inevitable sequencing bias in target capture, the resulting ITDs can only be qualitatively measured, and accurate quantitative results cannot be obtained.

Moreover, the clinical detection field is currently more inclined to use just one test on AML patients to obtain more detailed information and NGS could fully meet the requirement at this point. It needs only one blood test to complete multiple mutation detections of SNV, CNV, INDEL and ITD. Therefore, there is an urgent need to develop a new algorithm that enables the quantitative detection of FLT3-ITD based on the NGS platform.

SUMMARY OF THE INVENTION Technical problem to be solved by the present invention In view of the aforementioned problems in the prior art, the present invention develops a method for the quantitative detection of ITD mutation based on sequencing data --- using dual platforms (NGS platform and standard detection platform) to perform tests on a large number of collected ITD positive samples, i.e., using the PCR amplification-capillary electrophoresis as the gold standard to obtain the quantitative ITD mutation results, and based on the length of these detected ITDs to find their appropriate matches in the NGS data. Then through supervised machine learning, these samples are used as the training set to perform training, and finally a sequencing data-based model capable of directly predicting quantitative ITD mutation outcome is obtained, and the purpose of quantitatively detecting ITDs by next-generation sequencing is achieved. That is, based on a large number of samples with gold-standard ITD quantitative detection results, the present invention screens the characteristics related to ITD quantitative detection, and performs model training, thereby obtaining a model by which the accurate ITD quantitative detection result of the sample to be tested can be determined by using only the sequencing data and the ITD related characteristics. That is, the present invention includes: 1. An apparatus for the quantitative detection of ITD mutation based on sequencing data, including: a data acquisition module, which is used to acquire the sequencing data of samples to be tested; a data pre-processing module, which is connected to the data acquisition module, and is configured to extract characteristics of ITD of samples to be tested, wherein the ITD characteristics are the ITD characteristics of the whole region of nucleotide sequences or the ITD characteristics of specific regions of nucleotide sequences; a quantification module, which is connected to the data pre-processing module, and is configured to obtain the ITD mutation allele frequency based on the ITD characteristic coefficients and the ITD characteristics of samples to be tested; and a detection result output module, which is connected to the quantification module, and is configured to output the ITD mutation allele frequency as the quantitative detection result of the

ITD mutation.

In the present invention, the sequencing data may be fastq data obtained by converting the

NGS data acquired by sequencing with a high-throughput sequencer via the existing software. The

sequencing approach may be to first capture the target sequence in the exon region or other

specified regions of the sample, and then sequencing it (i.e., target sequence capture sequencing) by

using a high-throughput sequencer. Alternatively, whole-genome sequencing (WGS) of the sample

may also be feasible.

In the present invention, the ITD mutation may come from samples of species that usually

have ITD mutations. Preferably, samples are from mammals, more preferably humans. Specified

regions are usually selected for ITD mutation detection, such as internal tandem duplications (ITDs)

of the juxtamembrane region of the Fms-like tyrosine kinase 3 (FLT3) gene in patients with acute

myeloid leukemia (AML) (hg19, NCBI version 37; chr3:28608000-28608600)

In the present invention, the quantitative detection of a sample is the detection of variant allele

frequency (VAF) of the ITD mutation of the sample. The gold standard ITD detection method is

generally a method accepted by those skilled in the art and capable of accurately obtaining the VAF

of ITD mutation, the position of ITD mutation or the length of ITD mutation (including the total

length of the ITD and the length of the ITD repeats) etc., for example, the gold standard for current

ITD quantitative detection may be specific amplification by polymerase chain reaction (PCR) in

combination with capillary electrophoresis (CE) (PCR-CE) to detect VAF of ITD mutations.

2. The apparatus according to item 1, wherein the quantification module includes a quantitative

model sub-module for obtaining an ITD mutation allele frequency based on the ITD characteristic

coefficients and ITD characteristics of samples to be tested, wherein quantitative model set in the

quantitative model sub-module is represented by the following equation (1),

f(w, x) wO + w 1 x 1 + w 2x 2 + w 3 x 3 + --- + wnx...(1) in the equation (1), f(w, x) represents the ITD mutation allele frequency, won represent the

ITD characteristic coefficients, and xo-n represent the ITD characteristics.

3. The apparatus according to item 2, wherein the quantitative model is configured by a

coefficient training sub-module, and is configured to acquire the ITD characteristic coefficients WO-n,

and wherein the coefficient training sub-module includes:

a detection result acquisition unit, which is configured to acquire first detection results of first

test samples and second detection results of first test samples as a training set,

a machine learning unit, which is configured to use the first detection results of first test

samples and second detection results of first test samples as a training set, and obtain the ITD

characteristic coefficients wonthrough machine learning of the training set,

wherein,

the first detection results are the high-throughput sequencing data, and the second detection

results are the mutation allele frequency value obtained from the ITD standard detection, such as the

gold-standard ITD detection method.

4. The apparatus according to item 2, wherein the quantitative model is configured by a

coefficient training sub-module, and is connected to the data pre-processing module for acquiring

the ITD characteristic coefficients won, the coefficient training sub-module includes,

test samples and second detection results of first test samples, and acquire first detection results of

second test samples and second detection results of the second test samples,

characteristic coefficients wonby the machine learning of the training set,

a machine learning test unit, which is configured to perform tests using the first detection

results of second test samples, and then compare the mutation allele frequency value calculated by

the equation (1) with the second detection results of second test samples,

a test result assessment unit, which is configured to assess whether the result of comparison

meets the expectation,

a machine learning revise unit, which is configured to determine the values of the ITD

characteristic coefficients wonwhen the comparison result meets the expectation, and to modify

(such as increase, decrease or re-provide) the ITD characteristics xon adopted in the equation (1)

when the comparison result does not meet the expectation, and reset the ITD characteristic

coefficients won,

wherein,

gold-standard ITD detection method.

5. The apparatus according to item 3 or 4, wherein the assessment is to be made whether the

comparison result meets the expectation according to the following equation (2):

- - nsampes )2..(2 2 Xflsamples 1(y57,,

in the equation (2), yj represents the second detection results of second test samples, y9 represents the mutation allele frequency value calculated by equation (1), y7 represents the mean

value of the second detection results of the second test samples.

specifically, the concrete assessment method includes: setting an expected value (set value) for

R2 , the comparison result is concluded as meeting the requirement if R 2 is above the set value, and it is concluded that the comparison result does not meet the requirement if R 2 is less than the set

value. A preferred set value is 0.9.

6. The apparatus according to any one of items 1-5, wherein the ITD characteristic is selected

from one or two or more of the following characteristics: the position of the occurring ITD, the

length of the ITD, the nucleotide sequence characteristics of the ITD, the nucleotide sequence

characteristics before and after the position of the occurring ITD, and the nucleotide sequence

characteristics of a particular sequence.

In the apparatus of the present invention, the length of the ITD representing the ITD

characteristic may include, but is not limited to, the total length of the ITD segment or the length of

the repeating segment. Sequence characteristics representing a particular sequence may include, but

is not limited to, sequence complexity. Nucleotide sequence characteristics representing the ITD

characteristic may include, but is not limited to, sequence complexity or GC content, and the like.

Sequence complexity can be evaluated by using blast software (different parameters).

In the apparatus of the present invention, the number of ITD characteristics is not particularly limited, and the number of ITD characteristics that may be selected is, for example, 500 to 2000, and preferably, the number of the ITD characteristics is, for example, about 1,500.

7. A method for the quantitative ITD mutation detection based on sequencing data, including:

acquiring the sequencing data of samples to be tested;

extracting ITD characteristics of the sequencing data of samples to be tested, wherein the ITD

characteristics are the ITD characteristics of the whole region of the nucleic acid sequence or the

ITD characteristics of specific regions of nucleic acid sequences;

quantitatively detecting the ITD mutation allele frequency of samples to be tested, and

obtaining the ITD mutation allele frequency (the quantitative detection result) based on the ITD

characteristic coefficients and the ITD characteristics of samples to be tested.

8. The method according to item 7, wherein quantitative detection step is performed by a

quantitative model represented by the following equation (1),

f(w, x) wO + w 1 x 1 + w 2x 2 + w3 x 3 + -- + wx..(1) in the equation (1), f(w, x) represents the ITD mutation allele frequency, won represent the

ITD characteristic coefficients, and xon represent the ITD characteristics.

9. The method according to item 8, wherein the method for acquiring the ITD characteristic

coefficients won of the quantitative model includes:

acquiring first detection results of first test samples and second detection results of first test

samples as a training set,

obtaining the ITD characteristic coefficients won by the machine learning of the training set,

wherein,

the first detection result is the high-throughput sequencing data, and the second detection result

is the mutation allele frequency value of the ITD standard detection, such as the ITD gold-standard

detection method.

10. The method according to item 8, wherein the method for acquiring the ITD characteristic

coefficients won of the quantitative model includes:

samples, and acquiring first detection results of second test samples and second detection results of

second test samples,

using the first detection results of first test samples and the second detection results of first test sample as a training set, and obtaining the ITD characteristic coefficients won by the machine learning of the training set, the first detection results of second test samples are used for testing, and the mutation allele frequency value calculated by the equation (1) is compared with the second detection results of second test samples to assess whether the comparison result meets the expectation, if the comparison result meets the expectation, the ITD characteristic coefficients won are determined; if the comparison result does not meet the expectation, the ITD characteristics xon adopted in the equation (1) are modified (such as increased, decreased or re-provided), and the ITD characteristic coefficients won are reset, wherein, the first detection result is the high-throughput sequencing data, and the second detection result is the mutation ratio value of the ITD standard detection, such as the ITD gold-standard detection method.

11. The method according to item 9 or 10, wherein the assessment is to be made whether the

- - nsampes )2..(2 2 Xflsamples 1(y57,,

in the equation (2), yj represents the second detection results of the second test samples, y9 represents the mutation allele frequency value calculated by the equation (1), y7 represents the

mean value of the second detection results of second test samples.

12. The method according to any one of items 7-11, wherein the ITD characteristic is selected

characteristics before and after the position of the ITD, and the nucleotide sequence characteristics

of a particular sequence.

In the method of the present invention, the length of ITD representing the ITD characteristics

may include, but is not limited to, the total length of the ITD segment or the length of the repeating

segment. Sequence characteristic representing a particular sequence may include, but is not limited

to, sequence complexity. Nucleotide sequence characteristic representing the ITD characteristic may

include, but is not limited to, sequence complexity or GC content, and the like. Sequence complexity can be evaluated by using blast software (different parameters).

In the method of the present invention, the number of ITD characteristics is not particularly

limited, and the number of ITD characteristics that may be selected is, for example, 500 to 2000,

and preferably, the number of the ITD characteristics is, for example, about 1,500.

13. An electronic device, including:

a processor; and

a memory, in which the computer program instructions are stored, and when the computer

program instructions are executed by the processor, the method for the quantitative ITD mutation

detection based on sequencing data according to any one of items 7-12 is performed by the

processor.

BRIEF DESCRIPTION OF THE DRAWINGS

Various other advantages and benefits of the present application will become apparent to those

skilled in the art by reading the detailed description in the preferred embodiments below. The

drawings are only for the purpose of illustrating the preferred embodiments, and should not to be

considered as a limitation on this application.

Fig. 1 is a schematic diagram showing an apparatus for the quantitative ITD mutation detection

based on sequencing data according to an embodiment of the present application;

Fig. 2 is a schematic diagram showing a quantification module in the apparatus for the

quantitative ITD mutation detection based on sequencing data according to an embodiment of the

present application;

Fig. 3 is a schematic diagram showing a coefficient training sub-module in the apparatus for

the quantitative ITD mutation detection based on sequencing data according to an embodiment of

the present application;

Fig. 4 is a schematic diagram showing a coefficient training sub-module in the apparatus for

the present application;

Fig. 5 is a flow chart showing a method for the quantitative ITD mutation detection based on

sequencing data according to an embodiment of the present application;

Fig. 6 is a schematic diagram showing an electronic device according to an embodiment of the present application;

Fig. 7 is a graph showing the detection result according to a preferred embodiment of the

present application.

DETAILED DESCRIPTION OF THE INVENTION

The technical terms mentioned in the specification have the same meanings as those generally

understood by the skilled in the art, and if there is a conflict, the definition in the present

specification shall prevail.

In general, the terms used in this specification have the following meanings.

Machine learning: machine learning is a branch of artificial intelligence. The research of

artificial intelligence is a natural and clear thread from focusing on "reasoning" to focusing on

"knowledge", and then focusing on "learning". Obviously, machine learning is a way to realize

artificial intelligence, i.e., to solve problems in artificial intelligence by machine learning. In the

past 30 years, machine learning has developed into an inter-disciplinary subject involving many

areas, such as probability theory, statistics, approximation theory, convex analysis, and

computational complexity theory. Machine learning theory is primarily about designing and

analyzing algorithms that allow computers to automatically "learn". Machine learning algorithms

are a class of algorithms that automatically analyze data to obtain patterns from it and use them to

predict unknown data. Since the learning algorithms involve a large number of statistical theories,

machine learning is closely related to inductive statistics, also known as the statistical learning

theory. In terms of algorithm design, machine learning theory focuses on achievable, effective

learning algorithms. Many inference problems have the difficulty of no program to follow, so part

of the machine learning research is to develop the approximation algorithm that is easy to handle.

Machine learning has been widely used in the areas such as data mining, computer vision, natural

language processing, biometrics, search engines, medical diagnostics, detection of credit card fraud,

securities market analysis, DNA sequencing, speech and handwriting recognition, strategy games

and robotics.

Target sequence capture sequencing: is to customize genomic regions of interest into specific

probes and hybridize with genomic DNA in a sequence capture chip (or solution), after enriching

the DNA segments of target genomic regions, they are sequenced by using the next generation sequencing technology.

ITD: internal tandem duplication.

Summary of the application

As mentioned above, there is currently a need to quantitatively detect the ITD mutation based

solely on sequencing data, but accurate (being close or substantially consistent with the ITD

standard detection method) quantification (ITD mutation allele frequency) detection results cannot

be obtained based on sequencing data only through the existing software or algorithms. The ITD

standard detection method described in the present invention can be, such as gold-standard

detection method, for example, a method of PCR amplification-capillary electrophoresis, or other

commonly recognized detection methods capable of accurately obtaining the ITD mutation allele

frequency. Sequencing as described herein generally refers to the next generation sequencing, i.e.,

NGS sequencing.

The existing method for directly detecting the ITD mutation ratio by NGS sequencing, such as

PINDEL, is based on high-depth target region capture sequencing, and it is realized by using the

information from the captured target regions. Since the capture process may cause missing capture

of the segments where the ITD occurs, accurate quantitative results cannot be obtained.

The inventors of the present application found that, through collecting the characteristics

related to the ITD and the corresponding characteristic coefficients in the sequencing data, the

quantitative detection results of ITD mutations of samples to be detected can be nearly consistent

with the detection results of the gold-standard. The quantitative ITD mutation detection of samples

to be tested may employ, for example, the ITD-related characteristics and the corresponding

characteristic coefficients described in the present application, and the acquisition of the

ITD-related characteristics and corresponding characteristic coefficients may also employ, for

example, the machine learning method described herein.

Therefore, the basic idea of the present application is to solve the above technical problems,

and quantitatively determine the ITD mutations by obtaining the ITD related characteristics and the

corresponding characteristic coefficients.

Specifically, the present application provides an apparatus, a method, and an electronic device

for the quantitative ITD mutation detection based on sequencing data, wherein firstly acquiring sequencing data of samples to be tested, then extracting the ITD characteristics of the sequencing data of samples to be tested, and obtaining the ITD mutation allele frequency of samples to be tested based on ITD characteristic coefficients and ITD characteristics of samples to be tested, and wherein ITD characteristics are the ITD characteristics of the whole region of a nucleic acid sequence or the ITD characteristics of the specific region of a nucleic acid sequence;

Herein, those skilled in the art can understand that the apparatus, method and electronic device

for quantitatively detecting ITD mutation based on sequencing data provided by the present

application can be used for the quantitative ITD mutation detection of various sequencing data, for

example, the data of whole genome sequencing, and target sequence capture sequencing, etc., as

long as this sequencing method is commonly used in the current sequencing methods for detecting

ITD mutation. Therefore, even if the sequencing data of target sequence capture is mainly described

below as an example, the embodiments of the present application are not limited thereto.

After introduction of the basic principles of the present application, the exemplary

embodiments according to the present application will be described in detail with reference to the

accompanying drawings. It is apparent that the described embodiments are only a part of the

embodiments of the present application, and are not intended to show all embodiments. It should be

understood that the present application is not limited by the exemplary embodiments described

herein.

Exemplary Apparatus

based on sequencing data according to an embodiment of the present application. As shown in Fig.

1, an apparatus 1708 for the quantitative ITD mutation detection based on sequencing data

according to an embodiment of the present application includes:

a data acquisition module 100 for samples to be tested, which is configured to acquire the

sequencing data of samples to be tested;

a data pre-processing module 200 for samples, which is connected to the data acquisition

module, and is configured to extract the ITD characteristics of samples to be tested, wherein said

ITD characteristics are the ITD characteristics of the whole region of a nucleic acid sequence or the

ITD characteristics of the specific region of a nucleic acid sequence; a quantification module 300, which is connected to the data pre-processing module, and is configured to obtain ITD mutation allele frequency based on the ITD characteristic coefficients and the ITD characteristics of samples to be tested; and a detection result output module 400, which is connected to the quantification module, and is configured to output the ITD mutation allele frequency as a detection result of ITD mutation allele frequency of samples to be tested. In the quantification module 300, particularly in the embodiment of the present application as shown in Fig. 2, further includes: a quantitative model sub-module 310 for obtaining ITD mutation allele frequency based on the ITD characteristic coefficients and the ITD characteristics of samples to be tested, wherein quantitative model set in the quantitative model sub-module is represented by the following equation (1), f(w, x)=wO + w 1 x 1 + w 2x 2 + w 3 x 3 + --- + wnx...(1) in the equation (1), f(w, x) represents ITD mutation allele frequency, won represent ITD characteristic coefficients, and xon represent ITD characteristics. The quantitative model sub-module 310 is configured to acquire the ITD characteristic coefficients won, or to acquire the ITD characteristics xon and the ITD coefficients won corresponding thereto. Particularly, in the embodiment of the present application, as shown in Fig. 2, the ITD characteristics or the ITD characteristics and the ITD coefficients corresponding thereto of the quantitative model sub-module 310 are configured by a coefficient training sub-module 320. In the coefficient training sub-module 320, particularly, in a preferred example of the present invention, as shown in Fig. 3, a data acquisition unit 321 for a sequencing sample is configured to acquire the first detection results of first test samples and the second detection results of first test samples as a training set; and a machine learning unit 322 is configured to use the first detection results of first test samples and the second detection results of first test samples as a training set, and obtain the ITD characteristic coefficients won by the machine learning of the training set, wherein the first detection result is high-throughput sequencing data, and the second detection result is the mutation allele frequency value of ITD gold-standard detection method. In still another preferred example of the present invention, as shown in Fig. 4, a coefficient training sub-module 320 includes a data acquisition unit 323 for sequencing samples, which is configured to acquire the first detection results of first test samples and the second detection results of first test samples, and acquire the first detection results of second test samples and second detection results of second test samples; a machine learning unit 324 is configured to use the first detection result of first test samples and the second detection results of first test samples as a training set, and obtain the ITD characteristic coefficients won by the machine learning of the training set. A machine learning detection unit 325, which is configured to perform a test by using the first detection results of second test samples, and compare the mutation allele frequency value calculated by the equation (1) with the second detection results of second test samples; a test result assessment unit 326, which is configured to assess whether the comparison result meets the expectation; a machine learning revise unit 327, which is configured to determine values of ITD characteristic coefficients wonwhen the comparison result meets the expectation, and revise the ITD characteristics xo adopted in the equation (1) when the comparison result does not meet the expectation, re-providing (or re-selecting) and determining the ITD characteristic coefficients wo-n. Wherein the first detection result is high-throughput sequencing data, and the second detection result is the mutation allele frequency value of ITD gold-standard detection method. As described above, the apparatus 1708 for detecting ITD mutation allele frequency based on sequencing data according to examples of the present application can be used in various terminal devices, such as servers for targeted capture sequencing, and the like. In one example, the apparatus 1708 according to the present example can be integrated into a terminal device as a software module and/or hardware module. For example, the apparatus 1708 may be a software module in an operating system of the terminal device, or may be an application program developed for the terminal device; of course, the apparatus 1708 may also be one of a number of hardware modules of the terminal device. Alternatively, in another example, the apparatus 1708 for the quantitative ITD mutation detection based on sequencing data and the terminal device can also be separate devices, and the apparatus 1708 can be connected to the terminal device via a wired and/or wireless network, and the interactive information is transmitted according to the arranged data format.

Exemplary Method Fig. 5 is a flowchart showing a method for the quantitative ITD mutation detection based on sequencing data according to an embodiment of the present application. As shown in Fig. 5, a method for the quantitative ITD mutation detection based on sequencing data according to an embodiment of the present application includes: S100, acquiring the sequencing data of samples to be tested; S200, extracting ITD characteristics of the sequencing data of samples to be tested, wherein the ITD characteristics are the ITD characteristics of the whole region of a nucleic acid sequence or the ITD characteristics of the specific region of a nucleic acid sequence; S300, quantitatively detecting the ITD mutation allele frequency of samples to be tested, and obtaining the quantitative ITD mutation detection result of samples to be tested based on the ITD characteristic coefficients and the ITD characteristics of samples to be tested.

Exemplary Electronic Device

Hereinafter, an electronic device according to an embodiment of the present application will be

described with reference to Fig. 6.

Fig. 6 illustrates a block diagram of an electronic device according to an embodiment of the

present application.

As shown in Fig. 6, an electronic device 10 includes one or more processors 11 and memory

12.

The processor 11 may be a central processing unit (CPU) or other form of processing unit with

data processing capability and/or instruction executing capability, and may control other

components in the electronic device 10 to perform desired functions.

The memory 12 may include one or more computer program products, which may include

various forms of computer readable storage media, such as a volatile memory and/or a nonvolatile

memory. The volatile memory may include, for example, a random access memory (RAM), and/or

a cache, and the like. The nonvolatile memory may include, for example, a read only memory

(ROM), a hard disk, a flash memory, and the like. One or more computer program instructions can

be stored in the computer readable storage medium, and the processor 11 can execute the program

instructions to realize the method for the quantitative ITD mutation detection based on sequencing

data in each of the above embodiments according to the present application, and/or other desired

functions. Various contents such as the above-described ITD characteristics, ITD characteristic

coefficients, and the like can also be stored in the computer readable storage medium.

In one example, electronic device 10 may also include an input apparatus 13 and an output

apparatus 14 that are interconnected by a bus system and/or other form of connections (not shown).

For example, the input apparatus 13 can include, for example, a keyboard, a mouse, and the

like. The output apparatus 14 can output various kinds of information to the outside, such as the

detection result of the quantitative ITD mutation detection and the like. The output apparatus 14 can

include, for example, a display, a speaker, a printer, and a communication network and the remote

output apparatus connected thereto, and the like.

Of course, for simplicity, only some components of the electronic device 10 related to the

present application are shown in Fig. 6 and the components such as the bus, the input/output

interface, and the like are omitted. In addition to this, the electronic device 10 may also include any

other suitable components depending on the concrete cases of the applications.

Exemplary Computer Program Product and Computer Readable Storage Medium

In addition to the method and apparatus described above, the embodiment of the present

application can also be a computer program product including computer program instructions, when

the computer program instructions are executed by a processor, they make the processor perform

the steps of the method for the quantitative ITD mutation detection based on sequencing data

according to each of the embodiments of the present application described in the above section of

"exemplary method" in this specification.

As for the computer program product, any combination of one or more programming

languages can be used for writing the program codes for performing the operations of embodiments

of the present application, and the programming languages include object-oriented programming

languages, such as Java, C++, etc., and also include conventional procedural programming

languages, such as the "C" language, or similar programming languages. The program codes can be

executed entirely on the user's computing device, partially on the user's device, as a stand-alone

software package, partially on the user's computing device while partially on the remote computing

device, or entirely on the remote computing device or server.

Furthermore, embodiments of the present application can also be computer readable storage

medium with computer program instructions stored therein, when the computer program instructions are executed by a processor, they make the processor perform the steps of the method for the quantitative detection of ITD mutation based on sequencing data according to each of the embodiments of the present application described in the above section of "exemplary method" in this specification.

The computer readable storage medium can employ any combination of one or more readable

mediums. The readable medium may be a readable signal medium or a readable storage medium. A

readable storage medium can include, for example, but not limited to, an electronic, magnetic,

optical, electromagnetic, infrared, or semiconductor system, apparatus, or element, or any

combination of the above. More concrete examples (non-exhaustively listed) of readable storage

media include: an electrical connection with one or more wires, a portable disk, a hard disk, a

random access memory (RAM), a read only memory (ROM), an erasable programmable read-only

memory (EPROM, or flash memory), an optical fiber, a portable compact disk read only memory

(CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the

above.

Embodiments

Hereinafter, a concrete embodiment of the apparatus for the quantitative ITD mutation

detection based on sequencing data according to the present application will be described with

reference to Figs. 1-5, so as to explain the present invention in more details, but the present

invention is not limited by these embodiments.

First, the targeted capture sequencing data and PCR-CE ITD quantitative test results of the

FLT3 gene ITD frequently-occurring region (chrl3: 28608000-28608600) from 80 patients with

acute myeloid leukemia (AML) are collected and divided equally into ten groups, taking nine of

them (the first test samples) as the basis of a training set for establishing the quantitative model, and

the coefficient training sub-module is used to obtain the ITD characteristics and the ITD

characteristic coefficients corresponding thereto for the present embodiment. The remaining one

(the second test samples) is used to check whether the test result of the above quantitative model

meets the expectation. When the test result is concluded not to meet the expectation, the ITD

characteristics adopted by the quantitative model are revised (increased, decreased, or re-provided),

and the ITD characteristic coefficients are reprocessed.

Therefore, it can be understood that the quantification module of the embodiment includes a

quantitative model sub-module, wherein quantitative model set in the quantitative model

sub-module is represented by the following equation: f(w, x) wO + w 1 x 1 + w 2 x 2 + w 3 x 3

+ -- + w'x'.

wherein, f(w, x) represents ITD mutation allele frequency, won represent ITD characteristic

coefficients, and xon represent ITD characteristics.

In this embodiment, the quantitative model is configured by the coefficient training

sub-module, and is used to acquire the ITD characteristic coefficients won and the ITD

characteristics xon.

Particularly, in this embodiment, the coefficient training sub-module is configured with a

detection result acquisition unit, which is configured to acquire the first detection results of the first

test samples and the second detection results of the first test samples, and to acquire the first

detection result of the second test sample and the second detection result of the second test sample;

a machine learning unit, which is configured to use the first detection results of the first test samples

and the second detection results of the first test samples as a training set, and obtain the ITD

characteristic coefficients wo by the machine learning of the training set; a machine learning test

unit, which is configured to perform tests by using the first detection results of the second test

samples, and compare the mutation allele frequency value calculated by the quantitative model with

the second detection results of the second test samples; a test result assessment unit, which is

configured to assess whether the comparison result meets the expectation; a machine learning revise

unit, which is configured to determine the values of the ITD characteristic coefficients won when

the comparison result meets the expectation, and revise the ITD characteristics xon adopted by the

quantitative model when the comparison result does not meet the expectation, re-providing and

determining the ITD characteristic coefficients won; wherein the first detection result is

high-throughput sequencing data, and the second detection result is a mutation ratio value of ITD

standard detection.

The assessment of whether the comparison result meets the requirement is to be made by the

following equation (2):

1 _ R 2= 21 -nfsamples- -- 0sa ple - - 2...... (2) 1 ) ,nlsamples- in the equation (2), yi represents the second detection results of the second test samples, y, represents the mutation allele frequency value calculated by equation (1), y; represents the mean value of the second detection results of the second test samples.

The quantitative model, the ITD characteristics xon and the characteristic coefficients wo, of

the present embodiment are obtained by the above apparatus and operation.

Then, the targeted capture sequencing data (in the format of fastq file) and PCR-CE ITD

quantitative test results of the FLT3 genelTD frequently-occurring area (chr3: 28608000

28608600) from 30 patients with acute myeloid leukemia (AML) are collected.

And then the ITD characteristics of the sequencing data from above samples are extracted by

the data pre-processing module, wherein the ITD characteristics are the lTD characteristics of the

target capture area. These ITD characteristics include, but are not limited to: the length of the insert

segment (lTD mutant segment), the complexity of the insert segment (TD mutant segment), the

supporting sequencing reads number of the insert segment (lTD mutant segment), and the position

of the insert segment (ITD mutant segment), the depth of the insert segment (ITD mutant segment).

Finally, the detection result of the ITD mutation allele frequency is obtained by the quantitative

calculation module based on the ITD characteristic coefficients and the ITD characteristics of

samples to be tested. The detection result is output by an output module, as shown in the following

example,

chr13 28608251 21 TGAGATCATATTCANATTMc INS 0.0809866666666666

In the example of the displayed output detection result, the first and the second item are the

absolute position of the ITD occurring in the genome, the third item is the length of the ITD, the

fourth item is the sequence of the ITD, and the fifth term is the type of the ITD, and the final one is

the quantitative (lTD mutation allele frequency) detection result (the ITD quantitative result of this

example is 8.09%).

By using the apparatus and method for the quantitative ITD mutation detection based on the

sequencing data according to the present embodiment, the quantitative detection results of all 30

samples are shown in Fig. 7. In Fig. 7, the abscissa is the sample number, and the ordinate is the

lTD mutation allele frequency value. The model prediction curve is the ITD mutation allele

frequency value obtained by using the apparatus and method for detecting the lTD mutation allele

frequency based on sequencing data according to the preferred embodiment; and the NGS result curve is the ITD mutation allele frequency directly calculated without the model training, and the gold standard curve is the quantitative detection result by using the PCR-CE method. The R2 values of the model prediction curve and the NGS curve obtained by the equation (2) of the present embodiment are 0.9951 and 0.875 respectively. It can be seen from this, that the result of the ITD mutation allele frequency obtained by the apparatus and method for the quantitative ITD mutation detection based on sequencing data according to the preferred embodiment is more relevant to the gold standard and has a higher degree of conformity.

The fundamentals of the present application have been described above in conjunction with

particularly embodiments. However, it should be noted that the benefits, advantages, effects, and the

like mentioned in the present application are merely examples and not limitations, and the benefits,

advantages, effects, etc. are not considered to be required in each of the embodiments of the present

application. In addition, the specific details of the above disclosure are only for the purpose of

illustration and for ease of understanding, and are not intended to limit the present application. The

above details do not limit the application to be implemented by following the above specific details.

INDUSTRIAL APPLICABILITY

According to the present invention, provided are an apparatus and a method capable of

detecting an ITD mutation quantitatively while acquiring the ITD sequencing information.

Claims

1. An apparatus for the quantitative ITD mutation detection based on sequencing data,

including:

a data acquisition module, which is configured to acquire sequencing data of samples to be

tested;

a data pre-processing module, which is connected to the data acquisition module, and is

configured to extract ITD characteristics of samples to be tested, wherein the ITD characteristics are

ITD characteristics of the whole region of nucleotide sequences or ITD characteristics of specific

regions of nucleotide sequences;

a quantification module, which is connected to the data pre-processing module, and is

configured to obtain ITD mutation allele frequency based on the ITD characteristic coefficients and

ITD characteristics of samples to be tested; and

a detection result output module, which is connected to the quantification module, and is

configured to output the ITD mutation allele frequency as the quantitative detection result of the

ITD mutation of samples to be tested, wherein,

the quantification module includes a quantitative model sub-module for obtaining ITD

mutation allele frequency based on the ITD characteristic coefficients and ITD characteristics of

samples to be tested, wherein quantitative model set in the quantitative model sub-module is

represented by the following equation (1),

f(w, x) =wO + w 1 x 1 + w 2x 2 + w 3 x 3 + --- + wnx...(1) in the equation (1), f(w, x) represents ITD mutation allele frequency, won represent ITD

characteristic coefficients, and xon represent ITD characteristics;

wherein the ITD characteristic is selected from one or two or more of the following

characteristics: the position of the occurring ITD, the length of the ITD, the nucleotide sequence

characteristics of the ITD, the nucleotide sequence characteristics before and after the position of

the occurring ITD, and the nucleotide sequence characteristics of a particular sequence;

wherein the quantitative model is configured by a coefficient training sub-module, and is

connected to the data pre-processing module for acquiring the ITD characteristic coefficients won,

the coefficient training sub-module includes: a detection result acquisition unit, which is configured to acquire first detection results of first test samples and second detection results of first test samples, and acquire first detection results of second test samples and second detection results of second test samples, a machine learning unit, which is configured to use the first detection results of first test samples and second detection results of first test samples as a training set, and obtain the ITD characteristic coefficients wo by the machine learning of the training set, a machine learning test unit, which is configured to perform tests using the first detection results of second test samples, and compare the mutation allele frequency value calculated by the equation (1) with the second detection results of second test samples, a test result assessment unit, which is configured to assess whether the comparison result meets the expectation, a machine learning revise unit, which is configured to determine the values of ITD characteristic coefficients wonwhen the comparison result meets the expectation, and to modify the ITD characteristics xo adopted in the equation (1) when the comparison result does not meet the expectation, and reset the ITD characteristic coefficients won, wherein, the first detection results are high-throughput sequencing data, and the second detection results are the mutation allele frequency of the ITD standard detection.

2. The apparatus according to claim 1, wherein the assessment is to be made whether the comparison result meets the expectation according to the following equation (2):

2 nsampes-l (2 R 2= 1 - sa = ple - - 2...... (2) Xflsampesl1yy)

in the equation (2), yj represents the second detection results of second test samples, y9 represents the mutation allele frequency calculated by the equation (1), y7 represents the mean value of second detection results of second test samples.

3. A method for the quantitative ITD mutation detection based on sequencing data, including: acquiring sequencing data of samples to be tested; extracting ITD characteristics of sequencing data of samples to be tested, wherein the ITD characteristics are ITD characteristics of the whole region of the nucleic acid sequences or ITD characteristics of specific regions of nucleic acid sequences; quantitatively detecting the ITD mutation allele frequency of samples to be tested, and obtaining the quantitative detection result of samples to be tested based on ITD characteristic coefficients and ITD characteristics of samples to be tested, wherein, the quantitative detection step is performed by a quantitative model represented by the following equation (1), f(w, x)=wO + w 1 x 1 + w 2x 2 + w 3 x 3 + --- + wnx...(1) in the equation (1), f(w, x) represents the ITD mutation allele frequency, won represent the

ITD characteristic coefficients, and xon represent the ITD characteristics;

the occurring ITD, and the nucleotide sequence characteristics of particular sequences;

wherein the method for acquiring the ITD characteristic coefficients won of the quantitative

model includes:

second test samples,

using the first detection results of first test samples and second detection results of first test

samples as a training set, and obtaining the ITD characteristic coefficients won by the machine

learning of the training set,

the first detection results of second test samples are used for testing, and the mutation allele

frequency value calculated by the equation (1) is compared with the second detection results of

second test samples to assess whether the comparison result meets the expectation,

if the comparison result meets the expectation, the ITD characteristic coefficients won are

determined; if the comparison result does not meet the expectation, the ITD characteristics xon

adopted in the equation (1) are modified, and the ITD characteristic coefficients won are reset,

wherein,

is the mutation allele frequency of the ITD standard detection.

4. The method according to claim 3, wherein the assessment is to be made whether the comparison result meets the expectation according to the following equation (2):

R 1-_ fsamples-1 )2 . 2 (2) Xflsamples 1(y57,,

in the equation (2), yj represents the second detection results of second test samples, y9 represents the mutation allele frequency calculated by the equation (1), , represents the mean

value of the second detection results of second test samples.

5. An electronic device, including:

a processor; and

detection based on sequencing data according to claim 3 or 4 is performed by the processor.

1/ 5

2/ 5

3/ 5

4/ 5

5/ 5