Next generation sequencing applications for breast cancer research.

For some time, cancer has not been thought of as a disease, but as a multifaceted, heterogeneous complex of genotypic and phenotypic manifestations leading to tumorigenesis. Due to recent technological progress, the outcome of cancer patients can be greatly improved by introducing in clinical practice the advantages brought about by the development of next generation sequencing techniques. Biomedical suppliers have come up with various applications which medical researchers can use to characterize a patient's disease from molecular and genetic point of view in order to provide caregivers with rapid and relevant information to guide them in choosing the most appropriate course of treatment, with maximum efficiency and minimal side effects. Breast cancer, whose incidence has risen dramatically, is a good candidate for these novel diagnosis and therapeutic approaches, particularly when referring to specific sequencing panels which are designed to detect germline or somatic mutations in genes that are involved in breast cancer tumorigenesis and progression. Benchtop next generation sequencing machines are becoming a more common presence in the clinical setting, empowering physicians to better treat their patients, by offering early diagnosis alternatives, targeted remedies, and bringing medicine a step closer to achieving its ultimate goal, personalized therapy.

Background and aims

Science alone could never answer the great questions of humanity, nor could technology lead to improvement on its own. But when the two meet, progress can take place, and such is the case of discovering the molecular profile of phenotypic manifestations of various illnesses, which can eventually lead to the discovery of suitable and effective targeted therapies. Nowadays, molecular diagnosis for cancer patients prior to establishing the course of treatment is becoming a common presence in clinical practice, since nucleic acid sequencing has developed into a useful tool in the hands of medical researchers.

During the past years, the attention of healthcare practitioners and clinical researchers has begun to shift from treating patients already affected by various diseases to what is known as “predictive, preventive, and personalized medicine” (PPPM), and nowhere is this trend better illustrated than in the field of oncology [1]. For a while now, cancer has not been thought of as a disease, but a multifaceted, heterogeneous complex of genotypic and phenotypic manifestations that lead to tumorigenesis. Among tumors, breast cancer is the most frequent type of cancer diagnosed in women across the world, with an incidence of 22.9% according to GLOBOCAN-2008 [2], and the second most common type of cancer for both sexes combined, after lung cancer; at the same time, it is the principal cause of cancer death in women [2].

In the pursuit for the best response to therapy for each patient, caregivers are beginning to rely on more than the histopathological grading and staging of the disease and the general health of the patient. Due to recent advances in molecular and genetic research, clinicians now consider more specific information when deciding on therapeutic approaches: gene and molecular expression of tumors, mutation status of genes involved in a particular disease, polymorphisms or copy number variations. As an example, it is worth mentioning the increased capabilities of oncologists to predict the response to the recombinant monoclonal antibody trastuzumab (Herceptin) in HER2 positive breast cancer patients [3], or the response to cetuximab and panitumumab in mutated BRAF and/or KRAS colorectal adenocarcinoma patients [4]. The ability to foresee if a patient will respond to a therapeutic scheme based on their molecular profile does not only translate into potentially improving their outcome, but it also means an important step towards achieving the goal of personalized medicine.

As Bateman and Quackenbush stated in a 2009 editorial from the Bioinformatics Journal [5], the most outstanding outcome of sequencing and publishing the first reference human genome was not only knowing the genome itself, but all the technological advances that contributed to it, and the accelerated scientific progress derived from the Human Genome Project (HGP) [6]. Among the recent scientific and technical advances that impact medical research is next generation sequencing, which has the potential to improve clinicians’ approach on diagnosis and targeted therapies [7,8].

Leading suppliers of biomedical equipment have used these recent technological discoveries to produce various sequencing platforms based on the most recent advances in the field of next generation sequencing. Or, to be more precise, in the field of second generation sequencing, to better differentiate between Sanger sequencers and the more recent SMRT sequencing techniques (Single Molecule Real Time) developed by Pacific Biosciences [9], which are not the object of this paper. Many second generation sequencing platforms are available on the market, displaying a wide range of performances, reading capabilities, prices and possible applications, as seen in Table I [8,10-16] but they all need certain crucial steps which are summarized in Figure 1. These machines, which are becoming a familiar presence in molecular biology and clinical research laboratories, are produced and developed by the key players in the field, Roche (454 GS FLX, GS Junior), Illumina/Solexa (GAII, MiSeq, HiSeq) and Life Technologies (SOLiD, Ion Torrent, Ion Proton). The purpose of this paper is to provide a brief overview of some of the applications for breast cancer research developed for these platforms, as well as an outline of the sequencing workflow, focusing on the benchtop instruments which are more likely to be found in clinical research laboratories (GS Junior from Roche, MiSeq from Illumina, Ion Torrent from Life Technologies).

Table I

Brief characterization of some Second Generation Sequencing platforms.

Producer	Roche		Illumina/Solexa			Life Technologies
Platform	454 GS FLX	454 GS Junior	GAII	HISeq models	MiSeq	SOLiD	Ion Torrent	Ion Proton
Seq method	SBS Pyro	SBS Pyro	SBS RDT	SBS RDT	SBS RDT	SBS	SBS H+	SBS H+
Amplification	emPCR	emPCR	Bridge PCR	Bridge PCR	Bridge PCR	emPCR	emPCR	emPCR
Read length	400–700 bp*	400 bp	36–151 bp	up to 2×150 bp**	36–151 bp	35–75 bp	> 100 bp*****	up to 200 bp
Type of read	SE, PE	SE, PE	SE, PE	SE, PE	SE, PE	SE, PE	SE, PE	SE, PE
Output Gb/run	0.45 – 0.7	0.35	≤ 95	up to 600***	> 1	77–155****	> 1*****	up to 10
Run time	up to 24 h	10 h	2–14 days	27h to 11 days	4–27 h	2–7 days	2 h	2–4 h
Error rate (%)	1	1	≥0.1	≥0.1	>0.1	>0.01	~1
Error type	Indels	Indels	Substitutions	Substitutions	Substitutions	Substitutions	Indels
Cost of machine	up to $ 500k	Approx $115k	approx $ 540k	up to $740k	approx $128k	$525k	approx $80k	approx $150k
Pros	Longer reads, faster	Good read length, fast	Widely used, mature platform	Ultra high output, ease of use	Fully automated workflow	Ultra high output, scalable runs; can use 96 bar codes	Cheap, past, label free chemistry, highly scalable	WGS applications, 384 barcodes supported by the Torrent Suite Software
Cons	High cost per Mb, errors with homo polymer repeats	Like the FLX, plus lowest output among small sequencing instruments	High cost per Mb, low multiplexing capabilities	Less scalability, high overall costs	Insufficiently used	Shorter read length, long time for clonal preparation	Homopolymer errors, laborious (yet semi-automated) library preparation	Possibly high cost per Mb and errors

SBS = Sequencing by synthesis; SBL = Sequencing by ligation; RDT = Reverse dye terminator; SE = Single-ended read; PE = Paired-ended read

at the lower limit of the range for the Titanium

PE for the 1500 and 2500 on rapid run

for the 2000 and 2500 on high output

155 Gb for the XL model

for the 318 chip

Figure 1

Overview of the main steps in Next Generation Sequencing workflow.

Methods

454/Roche technology and workflow

454 Life Sciences, a biotechnology company from Connecticut, launched the first massively parallel DNA sequencer, the 454, probably the most widely used platform today [13,15]. After the first 454 machine became available on the market, it underwent several improvements regarding its performance and output, leading up to the benchtop version, the GS Junior. The technology employed by all 454 systems is sequencing by synthesis, and the actual sequencing is performed by pyrosequencing, in which a complex of enzymes and optical devices discriminate each pyrophosphate that is released during the attachment of nucleotides by the polymerase, but only if there are no extended homopolymer repeats [17-19]. The workflow is comparable to the one used when sequencing on the Ion Torrent. The input samples can be genomic DNA from BACs, plastids, bacteria, plants or animals for de novo sequencing, amplicons generated for targeted sequencing or re-sequencing, or cDNA.

The samples are first quantified and the DNA is randomly fragmented usually by nebulisation. After performing some end-repair procedures, adaptors are ligated to the double stranded DNA fragments. Two types of adaptors are used, A and B, the latter carrying a biotin group which contributes to the final separation of fragments that can be used for pyrosequencing. After ligation and denaturation, the mixture of fragments is exposed to streptavidin beads for library purification. Since the homozygous A-A fragments do not contain any biotin marker, they are flushed away. The homozygous B-B fragments are also eliminated, being constrained by the double biotin-streptavidin bond. What remains are the heterozygous fragments that constitute the DNA library which, prior to sequencing, needs to undergo an amplification step via emulsion PCR (emPCR) [7]. This is performed by using DNA capture beads on which each fragment is attached via the A adaptor at a one-to-one ratio by using precise and limiting dilutions. The water-in-oil emulsion is designed to trap each bead with a single DNA fragment attached, creating a high number of micro-reactors equipped with all the necessary PCR ingredients, where massive parallel amplification takes place. In this manner, hundreds of thousands of clonally amplified DNA fragments are generated on the surface of each capture bead [20]. After the water-in-oil emulsion is broken, the amplified library is placed on a fiber optic “MicroTiterPlate” containing millions of wells sized in such a way that each well is occupied by only one capture bead [21]. Enzyme beads are also added to each well, containing DNA polymerase, luciferase and ATP sulfurylase, necessary for recording each pyrophosphate released upon nucleotide incorporation.

During the following step, the sequencing per se, the plate is flushed with separate solutions containing each of the four nucleotides (dATP, dTTP, dGTP, dCTP) in a sequential manner and a precise, pre-defined order. As each nucleotide is incorporated into the chain based on complementarity, the pyrophosphate that is displaced during phosphodiester bond formation is converted by sulfurylase into ATP. During a subsequent reaction, luciferase oxidizes luciferin in the presence of ATP and oxygen, leading to the formation of a series of compounds and a quantum of light which is recorded by the CCD camera of the sequencer. If the template strand contains a homopolymer stretch, the intensity of the emitted light is directly proportional to the number of identical bases which were added to the new, complementary strand, but this may be a source of errors, since the light signal that is recorded is not always conclusive for the exact number of incorporated bases [22]. Other errors are prevented by flushing the PicoTiter plate with the enzyme apyrase between each run of nucleotides added to the reaction.

The raw data is processed and recorded in the form of a flowgram, and then the primary reads are assembled by overlapping multiple reads into longer strands to obtain consensus reads. The sequences are further mapped and compared against the known information from various databases.

Illumina/Solexa technology and workflow

Illumina uses a different approach for sequencing, the dye terminator technique, a method originally developed by the researchers from Manteia Predictive Medicine and from Solexa, a company bought by Illumina in 2007. Another particularity consists in the fact that the sequencing uses DNA libraries amplified through bridge-PCR. For this, the genomic DNA is randomly fragmented by sonication, the ends are repaired, and poly-A overhangs are added to the 3′ ends of each fragment by Klenow DNA polymerase exo(−) [9]. The poly-adenylated tails are used to add partially complementary specific adaptors to both ends of the double stranded fragments. The library is then size selected and placed on the flow cell for bridge-PCR amplification [7,15]. The flow cell used by Illumina is a solid support with the inner surface functionalized with two distinct oligonucleotide primers, forward and reverse, which are complementary to the two adaptors ligated on the DNA library [18]. The primers are attached with their 5′ end to the surface of the flow cell by a flexible linker, so that molecules that are amplified from a particular DNA fragment from the original library remain clustered together in groups of about 1,000 clonal amplicons. Prior to the actual sequencing, the amplicons are linearized and sequencing primers are hybridized to the 3′ ends. Then, a series of reagents is added to the flow cell for the “sequencing by synthesis” step.

Each round of interrogation is based on the elongation of the newly synthesized strand with only one nucleotide. The nucleotides used by Illumina platforms are modified in two ways. First, each of the four bases (A, T, C, G) is attached to a specific fluorochrome, so they do not need to be flushed over the flow cell separately [23]. The second modification consists in the fact that the bases resemble the dideoxynucleotides used in Sanger sequencing, but they are “reversible terminators”, since the 3′-OH is blocked by a cleavable fraction that prevents the incorporation of more than one base to the sequencing strand [8]. Therefore, each query is recorded independently, and the sequencing cannot continue until the terminator tag is released from the nucleotide to expose the hydroxyl group. This method overcomes the limitation of the 454 platforms regarding the biased output when incorporating homopolymer repeats. It also offers real-time feedback after each synthesis cycle, and provides faster results [17]. Nonetheless, sequencing with Illumina platforms is prone to some weaknesses, mainly concerning read length, and the reasons may be connected to incomplete and uneven removal of the terminating group. Still, the accuracy of reads makes Illumina the platform of choice in many laboratories, especially after the release of its bench top model, MiSeq, suitable for small genomes, exomes or targeted sequencing [11].

Ion Torrent technology and workflow

Ion Torrent, the “Personal Genome Machine” (PGMTM) released by Life Technologies in 2011, brings a new approach to second generation sequencing. Although the technology is still based on sequencing by synthesis, it does not require fluorescently labeled or chemiluminescent dNTPs, or an image-based detection of incorporated nucleotides. Instead, it uses a disposable chip built on the semiconductor technology to detect not the pyrophosphate released during nucleotide incorporation, but the proton that is also discharged during this process. In other words, the Ion Torrent chip acts as a pH-meter, detecting subtle pH shifts which occur when a phosphodiester bond is formed during elongation of the sequencing strand [23]. This novel technology, which does not require any optical devices and allows faster runs, together with its low price and automated possibilities for library preparation, make the Ion Torrent a suitable instrument for targeted sequencing or small genome analyses.

The workflow starts with the construction of DNA libraries. Genomic or cDNA is first fragmented into 200–400 bp strands, the ends are repaired, and the library is amplified and purified; then specific adaptors are attached to the ends of the fragments, and the gaps are filled in. When performing multiplex sequencing, such as sequencing DNA for more than one patient, specific barcodes are added to the fragments. The library is then size-selected and quality-controlled, either by using an Agilent Bioanalyzer or traditional quantitative PCR (qPCR). Next step is amplification via emulsion PCR [7]. After amplification, which resembles what we previously described for the 454 platform, the amplicon library undergoes template bead enrichment. For the PCR amplification and library enrichment, Ion Torrent developed the “Ion OneTouch” system, which automates the process and reduces hands-on time a great deal [12]. After this step, the DNA library is loaded on the chip and centrifuged, to ensure that each well of the chip is occupied by a template bead, and the sequencing process is initiated [15].

As previously mentioned, Ion Torrent technology is based on sequencing by synthesis, which consists in flooding the chip with individual, native nucleotides in a defined order. When complementary nucleotides are added to the sequencing strand by the DNA polymerase, the proton that is released causes a pH modification that is measured by sensors located at the bottom of each well. Since there are no optical devices involved, the sequencing reactions can proceed at a reasonably high rate [10,24]. Still, aside from the improvements brought by the Ion Torrent PGMTM, the problem of the homopolymer repeats still remains unsolved for the pyrosequencing based platforms, regardless of the way they detect nucleotide incorporation – as fluorescent signal or as a pH shift [14,25]. However, the release of the new Ion Proton and the rapid advancements in this field might lead to a solution to this limitation.

Results and discussion

NGS applications in breast cancer research

Second generation sequencing covers a wide range of applications, from whole genome sequencing (WGS) for individual patients or for other groups of organisms to exome sequencing, targeted re-sequencing and functional studies focused on the DNA – protein interactions (chromatin immunoprecipitation ChIP), which all have the potential to bring improvement in the quest for personalized medicine [26]. Human genetic variation applications developed for NGS include the evaluation of Single Nucleotide Polymorphisms (SNPs), Structural Variations (SVs) and Indels. The knowledge derived from these investigations can provide grounds for assessing possible links between genetic variations and different diseases, with high applicability in the field of oncology. Among these studies, the most common are genome-wide association studies (GWAS) and functional characterization of SNPs [16,27]. Many of these single nucleotide polymorphisms are connected to various pathological conditions either by abnormal gene expression or by altered protein structure and/or function, so they hold the potential to be used as genetic markers in the clinic to predict the risk of developing a certain disease, prognostic or response to therapy [28,29].

Breast cancer epidemic has reached alarming levels, so considerable efforts are made to learn more about this disease and the genetic alterations that lead to tumorigenesis, including cell growth, migration, angiogenesis and metastasis. Another aspect which requires attention is the lack of consistent and predictable response to main-stream chemotherapy used for mammary tumors classified according to traditional histopathological or other commonly used grading systems (TNM, Nottingham, etc). The reason for this is breast cancer heterogeneity – both inter-tumor and intra-tumor, which causes tumor-cell populations to behave differently and unpredictably [30-32].

The extensive knowledge unraveled at the end of the Human Genome Project enabled biochemical companies to produce commercial kits designed for the study of breast cancer, and cancer in general, with the help of NGS. Today, the market offers a wide range of cancer panels which can be used to discover the degree of tumor-specific genomic alterations by targeted re-sequencing of cancer genomes [13]. Some of the kits are produced by the manufacturers of next generation sequencers, to be used on their own machines – such as the IonAmpliseq Cancer Panel from Life Technologies, and the TruSeq Cancer Panel from Illumina, while others are generic and can be performed on many instruments. The cancer panels differ also with regard to the number of genes they contain and to the specificity of the results they produce. For instance, BRCA MASTR™ Dx from Multiplicom was designed in order to identify mutations in two tumor suppressor genes responsible for inherited breast and ovarian cancer, namely BRCA1 and BRCA2. On the other hand, the IonAmpliseq Comprehensive Cancer panel from Life Technologies can pinpoint mutations in 409 genes that are linked to the process of tumor formation and progression. Some information regarding the most frequently used commercial kits for NGS based studies in cancer research is presented in Table II [33-39].

Table II

Some commercially available cancer panel kits.

Company	Seq panel	Genes included	Targeted disease	Preferred seq machine
Ambry Genetics	BreastNext	ATM, BARD1, BRIP1, CDH1, CHEK2, MRE11A, MUTYH, NBN, PALB2, PTEN, RAD50, RAD51C, STK11, TP53	Mainly hereditary breast and/or ovarian cancer; some genes associated with other cancers	Any NGS platform
Ambry Genetics	OvaNext	ATM, BARD1, BRIP1, CDH1, CHEK2, EPCAM, MLH1, MRE11A, MSH2,SH6, MUTYH, NBN, PALB2, PMS2, PTEN, RAD50, RAD51C, STK11, TP53	Hereditary gynecological cancers (breast, ovarian, uterus)	Any NGS platform
Ambry Genetics	CancerNext	APC,ATM, BARD1, BRIP1, BMPR1A, CDH1, CHEK2, EPCAM, MLH, MRE11A, MSH2, MSH6, MUTYH, NBN, PALB2, PMS2, PTEN, RAD50, RAD51C,SMAD4, STK11, TP53	Breast, colon, ovarian, uterine, other	Any NGS platform
Multiplicom	BRCA MASTR™ Dx	BRCA1, BRCA2	Hereditary breast and/or ovarian cancer	Any NGS platform
Roche NimbleGen	NimbleGen Sequence Capture arrays	User-designed panels, according to the research aim	Any type of cancer, known to be related to a particular gene/group of genes	Roche 454/GS Junior
Ion Torrent/ Life Technologies	Ion AmpliSeq Cancer Hotspot Panel	ABL1, AKT1, ALK, APC, ATM, BRAF, CDH1, CDKN2A, CSF1R, CTNNB1, EGFR, ERBB2, ERBB4, FBXW7, FGFR1, FGFR2, FGFR3, FLT3, GNA11, GNAQ, GNAS, HNF1A, HRAS, IDH1, JAK2, JAK3, KDR, KIT, KRAS, MET, MLH1, MPL, NOTCH1, NPM1, NRAS, PDGFRA, PIK3CA, PTEN, PTPN11, RB1, RET, SMAD4, SMARCB1, SMO, SRC, STK11, TP53, VHL, EZH2, IDH2	It targets the hotspot regions of 50 oncogenes and tumor suppressor genes which have been related to various types of cancer	Ion Torrent PGM
Ion Torrent/ Life Technologies	Ion AmpliSeq Comprehensive Cancer Panel	409 oncogenes and tumor suppressor genes (including the 50 genes covered by the Cancer Hotspot Panel)	It targets genes frequently cited in tumorigenesis, as well as tumor suppressors	Ion Torrent PGM
Illumina	TruSeq Amplicon – Cancer Panel	ABL1, AKT1, ALK, APC, ATM, BRAF, CDH1, CDKN2A, CSF1R, CTNNB1, EGFR, ERBB2, ERBB4, FBXW7, FGFR1, FGFR2, FGFR3, FLT3, GNA11, GNAQ, GNAS, HNF1A, HRAS, IDH1, JAK2, JAK3, KDR, KIT, KRAS, MET, MLH1, MPL, NOTCH1, NPM1, NRAS, PDGFRA, PIK3CA, PTEN, PTPN11, RB1, RET, SMAD4, SMARCB1, SMO, SRC, STK11, TP53, VHL	Able to identify somatic mutations in a high number of mutational hotspots in cancer genomes from various samples	Illumina, including MiSeq
University of Washington	BROCA Cancer Risk Panel	APC, ATM, ATR, BABAM1, BAP1, BARD1, BMPR1A, BRCC36, BRIP1, CDH1, CDK4, CDKN2A, CHEK1, CHEK2, EPCAM, FAM175A (Abraxas), MLH1, MRE11A, MSH2, MSH6, MUTYH, NBN, PALB2, PMS2, PRSS1, PTEN, RAD50, RAD51, RAD51B, RAD51C, RAD51D, RBBP8, RET, SMAD4, STK11, TP53, TP53BP1, UIMC1, VHL, XRCC2, and XRCC3	Mainly hereditary breast and/or ovarian cancer; some genes associated with increased risk of other cancers	Any NGS platform

Five to ten percent of all mammary tumors are hereditary, and are connected to germline mutations in the genes BRCA1 and BRCA2 [40,41]. Still, only about 25% of familial breast cancers are associated with mutations in the two tumor suppressor genes [35,42]. Although scientists are far from pinpointing all genomic alterations that lead to this multifactorial, heterogeneous disease, some mutated genes were observed to take an active part in tumorigenesis, such as tumor suppressors (APC, CHEK 2, TP53, PTEN), DNA repair genes (NBN, MUTYH, MLH1, BRIP1), members of important signaling pathways implicated in cell division, proliferation and survival (ABL1, ATM, BRAF, FGFR1, JAK2, KRAS) or genes that encode for cell adhesion molecules (CDH1, CTNNB1, EPCAM) (See Supplementary Table I) [43]. These genes and others can be targeted and analyzed with the commercial cancer panels designed for next generation sequencing.

Conclusions

For the time being, the main purpose of these targeted re-sequencing studies is to assess risk or prognosis, but, with NGS platforms and reagents becoming more affordable and approaching the $1,000/genome threshold, sequencing can achieve more clinical applications [20]. Automation made benchtop NGS machines affordable also from the time perspective, since the results for targeted sequencing can be obtained within a work day, so they have the potential of becoming familiar presences in clinics. Thus, geneticists can provide caregivers rapid and essential information on the molecular profile of tumors, which can help in diagnosis, assessment of disease, and in planning the most appropriate treatment scheme, with maximum efficiency and minimal side effects [24,28]. Once this is accomplished, medicine will be one step closer to achieving its ultimate goal, personalized therapy. And when this aim is reached, chances are that next generation sequencing will have played a crucial role in this endeavor.

Supplementary Table I

Some of the genes involved in tumorigenesis, which are targeted in commercially available sequencing kits.

Gene	Location	Full name	UniGene	Function
				Native	When mutated
ABL1	9q34.1	c-abl oncogene 1, non-receptor tyrosine kinase	Hs.431048	Cell growth, survival, adhesion, differentiation, DNA damage response, apoptosis	Chronic myeloid leukemia (CML), when translocated with BCR - also found in AML and ALL
APC	5q21-q22	adenomatous polyposis coli	Hs.158932	Tumor suppressor, controls cell division	Colorectal cancer, Familial adenomatous polyposis
ATM	11q22.3	ataxia telangiectasia mutated	Hs.367437	Cell division, DNA repair, normal development of nervous and immune system	When homozygous, ataxia-telangiectasia. When heterozygous, breast cancer and others
BARD1	2q34-q35	BRCA1 associated RING domain 1	Hs.591642	Cell growth and division, together with BRCA1	Uncertain risk in breast cancer and neuroblastoma
BRAF	7q34	v-raf murine sarcoma viral oncogene homolog B1	Hs.550061	In RAS/MAPK pathway, role in differentiation, migration, apoptosis	Cardiofaciocutaneous and multiple lentigines syndromes. Oncogene, somatic mutations associated with many cancers
BRCA1	17q21	breast cancer 1, early onset	Hs.194143	Tumor suppressor, gene regulation	Breast cancer, ovarian cancer, pancreatic cancer
BRCA2	13q12.3	breast cancer 2, early onset	Hs.34012	Tumor suppressor, gene regulation	Breast cancer, ovarian, pancreatic, prostate. Also Fanconi anemia type D1 when homozygous
BRIP1	17q22.2	BRCA1 interacting protein C-terminal helicase 1	Hs.128903	DNA damage repair	Breast cancer when inherited heterozygous. Fanconi anemia when homozygous
CDH1	16q22.1	cadherin 1, type 1, E-cadherin (epithelial)	Hs.461086	Cell adhesion, cell signaling	Breast cancer, hereditary diffuse gastric cancer
CHEK2	22q12.1	checkpoint kinase 2	Hs.291363	Tumor suppressor, detection of DNA damage and strand breaks	Breast cancer, Li-Fraumeni syndrome, other cancers
CTNNB1	3p21	catenin (cadherin-associated protein), beta 1	Hs.476018	Cell adhesion, cell signaling	Pilomatricoma; colorectal, liver, ovarian cancer, medulloblastoma; desmoid fibromatosis
EPCAM	2p21	epithelial cell adhesion molecule	Hs.542050	Calcium independent cell adhesion molecule in normal epithelium and gastrointestinal carcinoma	Hereditary nonpolyposis colorectal cancer, congenital tufting enteropathy
ERBB2	17q12	Her-2/neu, v-erb-b2 erythroblastic leukemia viral oncogene homolog 2	Hs.446352	Encodes for growth factor receptors	Amplification found breast cancer, overexpression in receptors leads to aggressive forms of cancer
FGFR1	8p12	fibroblast growth factor receptor 1	Hs.264887	Cell division, regulation of cell growth and maturation	Alterations found in cancers via proliferation, migration, angiogenesis
HRAS	11p15.5	v-Ha-ras Harvey rat sarcoma viral oncogene homolog	Hs.37003	Regulates cell division through signal transduction	Oncogenes; involved in bladder, thyroid, kidney cancer; Costello syndrome
JAK2	9p24	Janus kinase 2	Hs.656213	Cell growth and proliferation, role in hematopoiesis	Leukemia, essential thrombocythemia, primary myelofibrosis
KRAS	12p12.1	v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog	Hs.505033	GTPase involved in cell division and differentiation	Oncogenes; involved in pancreatic, lung, and colorectal cancers; cardiofaciocutaneous and Noonan syndromes
MLH1	3p21.3	mutL homolog 1, colon cancer, nonpolyposis type 2	Hs.195364	DNA damage repair	Lynch syndrome; when homozygous, also causes leukemia and neurofibromatosis
MRE11A	11q21	MRE11 meiotic recombination 11 homolog A	Hs.192649	DNA double-strand break repair, telomere length maintenance, homologous recombination via exo- and endonuclease activity	Ataxia telangiectasia-like disorder, blocks meiotic recombination
MSH2	2p21	mutS homolog 2, colon cancer, nonpolyposis type 1	Hs.597656	DNA damage repair	Lynch syndrome; endometrium, stomach, intenstine, liver cancer, etc. Homozygous mutations can also cause leukemia or lymphoma
MSH3	5q11-q12	mutS homolog 3	Hs.280987	Heterodimer with MSH2, involved in post-replicative DNA mismatch repair system	Endometrial cancer
MSH6	2p16	mutS homolog 6	Hs.445052	Post-replicative DNA damage repair	Lynch syndrome; endometrium, stomach, intenstine, liver cancer, etc. Homozygous mutations can also cause leukemia or lymphoma
MUTYH	1p34.1	mutY homolog	Hs.271353	DNA damage repair by MYH glycosylase	Familial adenomatous polyposis
NRAS	1p13.2	neuroblastoma RAS viral (v-ras) oncogene homolog	Hs.486502	Cell division and differentiation	Noonan syndrome, melanoma, other types
NBN	8q21	nibrin	Hs.492208	DNA damage repair	Breast, ovarian, prostate cancer, melanoma, leukemia, Nijmegen breakage syndrome
PALB2	16p12.2	partner and localizer of BRCA2	Hs.444664	Interacts with BRCA2; tumor suppressor	Breast cancer when inherited heterozygous. Fanconi anemia when homozygous
PIK3CA	3q26.3	phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha	Hs.553498	Role in signaling cascades involved in cell growth, survival, proliferation, motility and morphology	Oncogene, in breast, colorectal, ovarian, liver, gastric, lung cancers
PTEN	10q23.3	phosphatase and tensin homolog	Hs.500466	Tumor suppressor, involved in apoptosis	Somatic mutations in prostate and endometrial cancer, glioblastoma, astrocytoma, melanoma
RAD50	5q31	RAD50 homolog	Hs.633509	DNA damage repair by holding the broken ends together during process	Suggested to contribute to breast cancer
RAD51C	17q25.1	RAD51 homolog C	Hs.412587	DNA damage repair and meiotic homologous recombination	Hereditary breast and ovarian cancer, Fanconi anemia
SMAD4	18q21.1	SMAD family member 4	Hs.75862	Controls gene activity and regulates cell proliferation	Colon, pancreas cancer; hereditary hemorrhagic telangiectasia, juvenile polyposis syndrome
SRC	20q12-q13	v-src sarcoma (Schmidt-Ruppin A-2) viral oncogene homolog	Hs.195659	Proto-oncogene with role in regulation of embryonic development and cell growth	Increased activity in colon carcinoma cells
STK11	19p13.3	serine/threonine kinase 11	Hs.515005	Tumor suppressor, role in tissue polarization and apoptosis	Breast cancer, non-small cell lung carcinoma, melanoma, pancreatic cancer; Peutz-Jeghers syndrome
TP53	17p13.1	tumor protein p53	Hs.437460	Tumor suppressor, “guardian of the genome” (Lane, 1992)	Breast, bladder, colorectal cancer, osteosarcoma, Li-Fraumeni syndrome