Family-based whole-exome sequencing identifies rare variants potentially related to cutaneous melanoma predisposition in Brazilian melanoma-prone families.

Genetic predisposition accounts for nearly 10% of all melanoma cases and has been associated with a dozen moderate- to high-penetrance genes, including CDKN2A, CDK4, POT1 and BAP1. However, in most melanoma-prone families, the genetic etiology of cancer predisposition remains undetermined. The goal of this study was to identify rare genomic variants associated with cutaneous melanoma susceptibility in melanoma-prone families. Whole-exome sequencing was performed in 2 affected individuals of 5 melanoma-prone families negative for mutations in CDKN2A and CDK4, the major cutaneous melanoma risk genes. A total of 288 rare coding variants shared by the affected relatives of each family were identified, including 7 loss-of-function variants. By performing in silico analyses of gene function, biological pathways, and variant pathogenicity prediction, we underscored the putative role of several genes for melanoma risk, including previously described genes such as MYO7A and WRN, as well as new putative candidates, such as SERPINB4, HRNR, and NOP10. In conclusion, our data revealed rare germline variants in melanoma-prone families contributing with a novel set of potential candidate genes to be further investigated in future studies.

1. Introduction

Ultraviolet radiation exposure is the leading environmental risk factor for the development of cutaneous melanoma [1]. Intermittent sun exposure and sunburns are highly associated with this type of skin cancer [2]. However, hereditary factors play important roles in melanoma etiology, although the genetic basis of melanoma susceptibility is complex and not fully understood [3].

Approximately 10% of all melanoma cases are caused by germline mutations, primarily affecting the p16 isoform of the CDKN2A gene, which is responsible for 20–40% of all hereditary melanoma cases [1, 4, 5]. More recently, other genes have been associated with familial melanoma, including BAP1, POT1, ACD, TERF2IP and POLE [6-8]. Altogether, mutations in these genes, associated with CDK4, TERT promoter, and MITF, are found in < 3% of melanoma-prone families in studied populations, and the majority (>70%) of familial cases are of unknown etiology [4, 8, 9].

Data regarding the prevalence of CDKN2A germline mutations in Brazilian patients fulfilling clinical criteria for familial melanoma are scarce; differ by geographic region and adopted diagnostic criteria; and disclose prevalence rates that vary from 4.5% to 14% [10-12]. In our previous study, CDKN2A germline mutations were detected in 14% of a cohort of 59 unrelated patients from the Southeast region of Brazil [12]. No CDK4 pathogenic variants have been identified in Brazilian melanoma-prone families to date [10, 12]. Only one patient with the MITF E318K variant was detected in 48 unrelated probands negative for CDKN2A variants [13]. Thus, in a significant number of Brazilian melanoma-prone families, no pathogenic variants have been identified, confounding the implementation of adjusted screening and management strategies.

Despite efforts to discover additional melanoma susceptibility genes by using genome-wide approaches such as genome-wide linkage analyses and exome sequencing, studies of either multiple melanoma-affected family members or large case-control cohorts have identified only a small number of candidate loci [14-16]. Thus, the aim of this study was to identify novel genomic variants potentially related to melanoma predisposition in melanoma-prone families. Consequently, we performed whole exome sequencing (WES) of 10 probands from 5 different families (2 probands/family) who developed melanoma and had previously tested negative for CDKN2A and CDK4.

2. Materials and methods

2.1. Patients

We selected 5 families with at least 2 cases of cutaneous melanoma among first-degree relatives, for a total of 10 individuals for WES. Selected patients belonged to melanoma-prone families receiving follow-up at the Familial Melanoma Clinic of the Skin Cancer Department and genetic counseling at the Oncogenetics Department at A.C. Camargo Cancer Center (ACC), São Paulo, Brazil. All diagnoses of melanoma were confirmed by histologic review of pathologic materials/reports or medical records. Eligible individuals were those who did not have detectable deleterious mutations in either CDKN2A or CDK4 genes [12]. The ten selected members of the five families were also screened for TERT promoter and MITF E318K variants [13] and for rare germline copy-number variations [17], with negative results for both analyses, as previously published. This study was conducted in compliance with the Declaration of Helsinki, and was approved by the Internal Ethics Committee Board of A.C.Camargo Cancer Center (#1728/12). All patients provided written informed consent.

2.2. Whole exome sequencing

Germline DNA was obtained from peripheral blood leukocytes, following the standard protocols of ACC Biobank. Briefly, DNA was extracted using the Puregene®-DNA purification Kit (Qiagen, Hilden, Germany) according to manufacturer’s instructions. DNA concentration, purity, and integrity were assessed by spectrophotometry (Nanodrop 2000—Thermo Fisher Scientific, Waltham, MA, USA) and fluorometry (Qubit—Life Technologies, Foster City, CA, USA). WES of all 10 patients was performed using the Ion Proton platform (Ion Torrent, Foster City, CA, USA). Genomic libraries were generated with the TargetSeq Exome Enrichment kit (Life Technologies, Foster City, CA, USA) and sequenced on an Ion Proton instrument using Ion PI Sequencing 200 Kit v3 and Ion PI Chip v3 (Thermo Fisher Scientific, Waltham, MA, USA), following the manufacturer’s protocol. The resulting sequences were mapped to the reference genome (GRCh37/hg19). Base calling and alignment were performed by using a Torrent Suite v4.2 server and TMAP software (Torrent Mapper 4.2.18). Genomic variant calling was performed in two steps: (1) using the TVC 4.0–5 software (Torrent Variant Caller) following the Ion Torrent protocol (http://mendel.iontorrent.com/ion-docs/); (2) validation of variants using the GATK pipeline (https://www.broadinstitute.org/gatk/guide/best-practices?bpm=DNAseq). The comparison between the number of variants called by each pipeline and the numbers of concordant calls are described in the S1 Table. The exome sequencing data obtained in this study are available at Sequence Read Archive (PRJNA705160).

2.3. WES variant prioritization

Variant annotation was performed using public databases: dbNSFP (http://sites.google.com/site/jpopgen/dbNSFP) version 2.4; COSMIC v69; 1000 genomes; Exome Variant Server (http://evs.gs.washington.edu/EVS/) version ESP6500SI-V2; HapMap; and dbSNP version 138 through the SnpEff software version 3.5 using an in-house script developed by the ACC Bioinformatics Department. Variants detected in all 10 samples were disregarded because they could be sequencing artifacts or polymorphisms of the Brazilian population. In addition, our data were compared against an independent set of 20 exomes from Brazilian non-cancer patients (collaboration with the Human Genetics Lab–Dr. Krepischi–Institute of Biosciences, University of São Paulo), and all variants detected in this additional set were also excluded.

Variant prioritization was performed using VarSeq software (Golden Helix), with the following criteria: depth coverage >20 reads; Phred score >20; allelic frequency > 0.2; population frequency <1% (according to NHLBI GO Exome Sequencing Project, the 1000 Genomes Project, the Exome Aggregation Consortium [ExAC], the Online Archive of Brazilian Mutations [ABraOM], and the dbSNP 147). Variants were then selected according to their predicted impact on protein expression: loss of function (frameshift, nonsense, initiator codon alteration, and splice acceptor/donor variants) and missense, including inframe deletions/insertions. Finally, only variants present in both patients from the same family were analyzed further.

2.4 Targeted Next-Generation Sequencing (NGS) Validation

A subset of 66 variants selected from exome data were validated by multiplex targeted NGS with a custom Ion AmpliSeq panel. Libraries were prepared with 20 ng of DNA from each patient using an Ion AmpliSeq™ Library Kit 2.0 (Life Technologies), and sequencing was performed using the Ion Proton platform according to the manufacturer’s instructions. Sequencing reads mapped to the human genome reference (hg19) using Torrent Suite Browser 4.0.1, and variants were identified using the VariantCaller v4.0.r73742 plugin, considering as criteria for variant calling a base coverage ≥10x and VAF > 20%.

2.5. Gene pathway analysis and in silico prediction

We also used gene and pathway analysis software (Ingenuity Pathway Analysis [IPA] and VarElect [http://varelect.genecards.org/]) and in silico pathogenicity prediction software to identify representative pathway networks and to pinpoint other genes that may be important to melanoma susceptibility. The in silico pathogenicity prediction scores from SIFT, Polyphen2, LRT, Mutation Taster, Mutation Assessor and FATHMM/MKL software were annotated using VarSeq software (Golden Helix).

3. Results

The pedigrees of the 5 investigated families are presented in Fig 1A–1E. Clinical data from all 10 studied individuals are provided in Table 1. The mean age at diagnosis was 40 years old (range 18–65 years old). Half of the patients were diagnosed at age 40 or younger, and most (80%) diagnoses occurred before the sixth decade of life. Six (6/10) patients presented with a single cutaneous melanoma, while 2 cases had multiple lesions (>2). Most patients showed Fitzpatrick phototype I or II (7/10) and nevi counts >50 (6/10). Only 3 cases showed the atypical mole syndrome phenotype. Thyroid cancer and non-melanoma skin cancer where the most prevalent second neoplasms (two cases each) (Table 1).

Fig 1

Pedigrees of the five melanoma-prone families.

Tumor types are described beneath each individual, followed by the age of onset. Small black arrow indicates the index case of each family. The green arrow indicates the individuals who were subjected to WES analysis.

Table 1

Clinical characteristics of cutaneous melanoma patients.

Clinical Aspects	Family 1		Family 2		Family 3		Family 4		Family 5
	A	B	A	B	A	B	A	B	A	B
Gender	Female	Male	Female	Female	Female	Female	Female	Male	Male	Female
Age at melanoma diagnosis	51	65	30	27	51	46	44	18	39	35
Number of melanomas	2	8	1	3	1	1	1	1	2	1
Other cancer	Colon	Prostate Thyroid BCC	Thyroid	-	-	-	BCC	-	BCC	-
Kindred	Sister	Brother	Mother	Daughter	Sister	Sister	Mother	Son	Son	Mother
Other cancers in family	Melanoma, leukemia		BCC, breast		Melanoma, liver, endometrium		Lymphoma, kidney		Lung, BCC, SCC
Phototype	I	II	III	III	II	III	II	II	I	I
Hair color	Brown	Brown	Brown	Brown	Blond	Brown	Brown	Blond	Blond	Blond
Eye color	Brown	Brown	Brown	Brown	Blue	Brown	Brown	Brown	Blue	Blue
Nevi count	100–150	>150	>150	100–150	<50	50–100	< 50	50–100	< 50	< 50
Atypical Mole Syndrome	No	Yes	Yes	No	No	Yes	No	No	No	No

BCC = basal cell carcinoma; SCC = squamous cell carcinoma of the skin

3.1. WES variant prioritization for identifying melanoma predisposing genes

An average of 45,899,246 sequence reads was obtained for each patient and an average of 86% of the target bases was covered more than 20X. First, we used the WES data to investigate variants affecting 10 genes previously associated with melanoma predisposition (CDKN2A, CDK4, BAP1, POT1, TERT, ACD, TERF2IP, POLE, MITF, and MC1R), and classified them according to the American College of Medical Genetics and Genomics guidelines [18]. Except for risk alleles in MC1R, we did not find any pathogenic/likely pathogenic variants or variants of uncertain significance in these predisposition genes. In three families, one or both relatives harbored MC1R variants previously associated with increased melanoma risk (low or high-risk variants) (Table 2).

Table 2

Clinical characteristics of melanoma patients.

Family	Individual	MC1R Variants (zygosity)	dbSNP/ ABraOM MAF	Risk classification*
3	A	p.Val60Leu (ht); p.Arg160Trp (ht)	rs1805005/ 9.8%; rs1805008/ 2.2%	r; R
	B	p.Val60Leu (ht)	rs1805005/ 9.8%	r
4	A	p.Arg160Trp (ht)	rs1805008/ 2.2%	R
	B	p.Val60Leu (ht)	rs1805005/ 9.8%	r
5	A	none	-	-
	B	p.Val92Met (ht); p.Thr314 = (ht)	rs2228479/ 3.7%; rs2228478/ 14.5%	r; r

*R: variants associated with red hair color and more than 2X increased risk for melanoma; r: variants not associated with red color hair and 1-2X increased risk for melanoma [19]. ht: heterozygous. ABraOM: database of Brazilian genomic variants obtained with whole-exome and whole-genome sequencing from 1,171 unrelated individuals (http://abraom.ib.usp.br/index.php). MAF: minor allele frequency.

To prioritize variants in other genes, we applied several filters focusing on quality, frequency, and effect of the identified variants and their occurrence in both affected relatives for each family (Fig 2). A total of 288 heterozygous rare non-synonymous variants in 281 genes were identified that co-segregated in both relatives of each family, with 281 missense and 7 loss-of-function (LoF) variants (Table 3 and S2 Table). All variants were exclusive for one given family, and only one gene had different prioritized variants in two families (UNC93A gene–Family 1 variant p.Arg226Ter and Family 4 variant p.Gly152Asp). Seven genes harbored rare LoF variants (3 frameshift and 4 nonsense) detected in three families (ADGRG7, FAM221A, SERPINB4, UNC93A, HRNR, OR51M1, SLC5A11) (Table 4). We also performed a technical validation of the prioritized variants, selecting a subset of these 288 variants (66 out of 288) for targeted NGS in the same WES samples, and all were validated (S3 Table).

Fig 2

Diagram of the germline sequencing analysis and variant prioritization strategy, showing the main technical processes.

WES data from 10 melanoma patients were analyzed using quality and frequency-based filters, resulting in 288 rare non-synonymous variants identified in both affected members of each family, which were investigated further for their predicted pathogenicity and gene function. WES, whole exome sequencing. TMAP, Torrent Mapper; TVC, Torrent Variant Caller; GATK, Genomic Analysis Toolkit; QC, Quality Control; MAF, minor allele frequency; LoF, Loss of Function.

Table 3

Types of rare non-synonymous heterozygous variants co-segregating in both affected relatives of the five melanoma-prone families.

	Family 1	Family 2	Family 3	Family 4	Family 5	Total
Total of variants	77	57	72	51	31	288
Missense	73	57	71	49	31	281
LoF variants
Nonsense	1	0	1	2	0	4
Frameshift	3	0	0	0	0	3

LoF: Loss-of-Function

Table 4

Rare germline LoF variants identified by WES in the 5 melanoma-prone families.

Family	Gene	dbSNP id	Genomic position (Hg19)	Type	Exon	RefSeq	c. HGVS	p. HGVS	MAF ExAC	MAF ABraOM
1	ADGRG7	rs574492402	3:100378552	Frameshift	14	NM_032787	c.1843_1844insA	p.Pro616Thrfs	0.0003298	0.001642
	FAM221A	rs553824715	7:23731209	Frameshift	3,4	NM_199136	c.631delG	p.Ile212Leufs	n/d	0.000821
	SERPINB4	rs554627371	18:61306960	Frameshift	6	NM_002974	c.520delC	p.Leu174Trpfs	0.000008288	0.004105
	UNC93A	rs145360877	6:167717457	Stop codon	4,5	NM_018974	c.676C>T	p.Arg226Ter	0.002356	0.004926
3	HRNR	rs141263661	1:152191578	Stop codon	3	NM_001009931	c.2527C>T	p.Arg843Ter	0.0002224	n/d
4	OR51M1	rs182074434	11:5410769	Stop codon	2	NM_001004756	c.141C>G	p.Tyr47Ter	0.001582	0.000821
	SLC5A11	rs147549055	16:24873990	Stop codon	3,4	NM_052944	c.204G>A	p.Trp68Ter	0.0007087	n/d

n/d–not described. MAF–minor allele frequency.

A total of 281 genes were encompassed by the 288 variants. To evaluate the molecular mechanisms and potential roles of these genes in pathogenesis and clinical phenotypes, we performed an analysis using the VarElect tool (http://varelect.genecards.org/), which associates genes and phenotypes based on shared pathways, interaction networks, paralogy relationships, and mutual publications. We used the following terms to perform this analysis: cancer; cancer susceptibility; melanoma; melanoma susceptibility; skin pigmentation; melanocyte; melanosome; DNA repair; cell cycle; and telomeres. The most relevant genes (top 20 genes with the highest connection scores) associated with the terms cited above are shown in Table 5; all genes and scores provided by this analysis are provided in S4 Table.

Table 5

Detected genes associated with the phenotypes of interest and their respective rare non-synonymous variants.

Genes	Matched Phenotypes	Matched Phenotypes Count	Score	Average Disease Causing Likelihood (%)	Family	SNP id	HGVS c.	HGVS p.	MAF ExAC
FANCA	Cancer, cancer susceptibility, melanoma, skin pigmentation, melanocyte, DNA repair, cell cycle, telomeres	8	164	32%	2	rs17233141	c.2574C>G	p.Ser858Arg	0.01
WRN	Cancer, cancer susceptibility, melanoma, skin pigmentation, melanocyte, DNA repair, cell cycle, telomeres	8	146	14%	3	rs4987238	c.1149G>T	p.Leu383Phe	0.001919
WRN	Cancer, cancer susceptibility, melanoma, skin pigmentation, melanocyte, DNA repair, cell cycle, telomeres	8	146	14%	3	rs140768346	c.2983G>A	p.Ala995Thr	0.002207
TYMP	Cancer, cancer susceptibility, melanoma, skin pigmentation, melanocyte, DNA repair, cell cycle, telomeres	8	70	35%	3	rs143789597	c.242G>A	p.Arg81Gln	0.001112
NOP10	Cancer, skin pigmentation, melanocyte, DNA repair, cell cycle, telomeres	6	52	66%	1	rs146261631	c.34G>C	p.Asp12His	0.009744
PTPN22	Cancer, cancer susceptibility, melanoma, skin pigmentation, melanocyte, cell cycle, telomeres	7	40	29%	1	rs72650671	c.1108C>A	p.His370Asn	0.002273
MCM3	Cancer, cancer susceptibility, melanoma, melanocyte, DNA repair, cell cycle, telomeres	7	35	49%	3	rs148636199	c.1618C>T	p.Arg540Trp	0.00007413
RECK	Cancer, cancer susceptibility, melanoma, melanocyte, DNA repair, cell cycle, telomeres	7	33	48%	1	rs375477269	c.1747G>A	p.Val583Ile	0.00004942
MUC16	Cancer, cancer susceptibility, melanoma, DNA repair, cell cycle	5	32	0%	1	rs184811119	c.14885C>T	p.Thr4962Ile	0.003101
MTUS1	Cancer, cancer susceptibility, melanocyte, DNA repair, cell cycle, telomeres	6	31	10%	4	rs61733691	c.1936G>C	p.Glu646Gln	0.003923
KMT2D	Cancer, skin pigmentation, melanocyte, DNA repair, cell cycle	5	26	61%	1	rs189888707	c.7670C>T	p.Pro2557Leu	0.00834
LRRC56	Cancer, melanoma, skin pigmentation, melanocyte	4	25	15%	1	rs61736743	c.544C>A	p.Gln182Lys	0.005785
LRRC56	Cancer, melanoma, skin pigmentation, melanocyte	4	25	15%	1	rs138291757	c.655G>A	p.Val219Met	0.002642
HPS5	Cancer, melanoma, skin pigmentation, melanocyte, DNA repair, cell cycle	6	24	16%	2	rs143784823	c.1501G>A	p.Gly501Arg	0.004406
ITGA3	Cancer, cancer susceptibility, melanoma, melanocyte, DNA repair, cell cycle	6	24	31%	1	rs140248487	c.2501C>T	p.Thr834Met	0.0003789
LCN2	Cancer, cancer susceptibility, melanoma, melanocyte, cell cycle, telomeres	6	23	42%	5	rs147787222	c.26G>T	p.Gly9Val	0.001367
ECM1	Cancer, cancer susceptibility, melanoma, melanocyte, DNA repair, cell cycle	6	23	24%	1	rs151102225	c.1181A>T	p.Asp394Val	0.007734
DST	Cancer, melanoma, skin pigmentation, melanocyte, DNA repair, cell cycle, telomeres	7	20	50%	1	rs138967674	c.7463C>A	p.Pro2488His	0.00722
TTN	Cancer, melanoma, DNA repair, cell cycle, telomeres	5	20	0%	3	rs72648244	c.91573A>G	p.Ile30525Val	0.00679
SELP	Cancer, cancer susceptibility, melanoma, skin pigmentation, melanocyte, DNA repair, cell cycle, telomeres	8	20	14%	1	rs144853111	c.2180G>A	p.Gly727Glu	0.001211
ADH1B	Cancer, cancer susceptibility, melanoma, DNA repair, cell cycle, telomeres	6	19	20%	3	rs6413413	c.178A>T	p.Thr60Ser	0.006589
LAMC1	Cancer, cancer susceptibility, melanoma, melanocyte, DNA repair, cell cycle	6	18	37%	1	rs34995260	c.3796G>A	p.Glu1266Lys	0.003468

MAF–minor allele frequency.

Among the identified rare germline variants, we selected LoF or missense alterations considered deleterious to protein function according to at least four of six prediction algorithms which forecast potential protein malfunctions: SIFT, Polyphen2, LRT, Mutation Taster, Mutation Assessor and FATHMM/MKL. Thus, we obtained a set of 91 rare germline variants affecting 91 different genes and potentially affecting protein function (S5 Table).

3.2. Gene pathway analysis

To identify representative pathway networks associated to the 91 genes affected by non-synonymous rare germline variants predicted to affect protein function, we performed an analysis using the Ingenuity Pathway Analysis software (IPA). This set was related principally to cancer, dermatological diseases and conditions, organismal injury and abnormalities, developmental disorders, and hereditary diseases. Furthermore, this classification was associated predominantly with skin cancer, melanoma, and tissue tumorigenesis (p-value <0.00004 –S6 Table).

The main network which comprised 35 genes of the initial set was associated primarily with cancer, organismal injury and abnormalities, and cellular growth and proliferation. These consisted of 13 genes that harbor rare germline variants in melanoma-prone families, and an additional 22 genes that were automatically included in the network because they have been biologically linked to 13 genes implicated by scientific evidence (Fig 3). The category, function/disease, and genes associated to the network are described in Table 6.

Fig 3

Network of 13 genes harboring rare germline variants identified in melanoma-prone patients.

Network generated by the IPA software displaying interactions between 13 genes identified by exome sequencing and 22 other genes automatically included after they were identified as biologically connected based on scientific evidence. The functional categorization of this network was cancer, dermatological diseases, and organismal injuries and abnormalities. The red nodes represent genes identified by this study harboring missense variants; the green nodes represent genes identified harboring LoF variants; the white node genes were plotted by the software once they are associated by scientific evidence. The nodes highlighted in pink represent genes involved in melanoma tumorigenesis according to IPA.

Table 6

Main cellular function and diseases associated with 91 prioritized genes.

Categories	Diseases or Functions Annotation	Molecules	p-Value
Cancer, Organismal Injury and Abnormalities	Connective tissue tumor	CACNA1S, CCND1, DSC3, DST, GDF2, HRNR, IGF1, IL4, MAD1L1, MCM3, STEAP4, TERT, TLE2, TNF	8.07E-08
Cellular Growth and Proliferation, Tissue Development	Proliferation of epithelial cells	CASP3, CCND1, Cdk, EP300, GDF2, GRN, IGF1, IL4, KIF3A, PKP3, STEAP4, TERT, TNF	9.46E-08
Cancer, Organismal Injury and Abnormalities	Neoplasia of cells	CACNA1S, CASP3, CCND1, DSC3, DST, EP300, GDF2, GLI3, GRN, GZMM, HRNR, IGF1, IL4, KIF3A, MAD1L1, MGEA5, MYO7A, OBSCN, SCGB1A1, SERPINB4, STEAP4, TERT, TLE2, TNF	2.85E-07
Embryonic Development, Organ Development, Organismal Development, Tissue Development	Development of sensory organs	CASP3, CCND1, CEP290, GLI3, GRN, IGF1, IL4, LYVE1, MYO7A, PAX8, TNF	5.41E-07
DNA Replication, Recombination, and Repair	DNA metabolism	CASP3, EP300, GDF2, GRN, GZMM, IGF1, IL4, MCM3, MGEA5, TNF	5.46E-07
Cell Morphology, Cellular Function and Maintenance	Mitochondrial transmembrane potential	CASP3, CCND1, EP300, GZMM, IL4, MGEA5, TERT, TNF	7.09E-07

Genes in bold are those with LoF variants.

4. Discussion

Whole genome and whole exome sequencing technologies are powerful tools to identify new cancer predisposition genes. These methods have been applied recently to discover melanoma predisposition genes such as MITF, TERT and POT1 [6, 20, 21]. However, despite the description of nearly a dozen melanoma predisposition genes, genetic etiology remains unknown in almost 80% of all melanoma-prone families [9].

In this study, we utilized WES to identify rare germline variants shared between first-degree relatives with cutaneous melanoma from five families, and to discover variants contributing to melanoma susceptibility. By applying several filtering strategies, we found 7 LoF variants in three families, and 91 variants predicted to impair gene function by in silico analysis. The seven genes affected by LoF variants–ADGRG7, FAM221A, SERPINB4, UNC93A, HRNR, OR51M1, and SLC5A11 –are not associated with genetic diseases according to the OMIM database and have not been related to melanoma susceptibility previously. Nevertheless, SERPINB4 and HRNR appeared in the “Neoplasia of cells” list identified by the IPA software evaluation of gene pathways of interest.

SERPINB4 encodes squamous cell carcinoma antigen 2, a member of the serpin family that has serine protease inhibitor functions, and that was initially discovered as a tumor-specific antigen in uterine carcinomas and later described as a biomarker for inflammatory skin diseases [22]. Recently, somatic mutations in SERPINB4 and SERPINB3 (predominantly missense mutations) were described in melanoma, and were associated with improved survival after anti-CTLA4 immunotherapy [23]. The second gene, HRNR, encodes hornerin, an epidermal protein first described in psoriatic lesions and in cutaneous wound healing [24]. Makino et al. have also shown in a murine model xenotransplanted with human skin that ultraviolet B (UVB) irradiation induces hornerin expression, leading to epidermal hyperproliferation and probably to tissue repair after UVB-induced injury [25]. Hornerin was recently shown to be highly expressed by pancreatic tumor endothelium; to alter tumor vessel parameters through a VEGF-independent mechanism [26]; and to promote tumor progression in human tissues and in cell models of hepatocellular carcinoma [27].

We also used gene and pathway analysis software (VarElect and IPA) and in silico pathogenicity prediction software to pinpoint other genes that may be important to melanoma susceptibility. From our list of prioritized variants and genes, three genes (MYO7A, WRN and NOP10) warrant a more detailed discussion, due to gene function and previous associations with melanoma.

A rare missense variant in MYO7A was identified in Family 2. Myosin has an essential role in melanosome transport and distribution [28]. In an analysis using IPA software, MYO7A was associated with melanoma, cancer, and melanosome degradation and localization. Gibbs et al. found that the absence of MYO7A expression in murine retinal pigmented epithelium impaired melanosome motility, thereby impeding the peripheral localization of melanosomes in melanocytes [29], thus showing the importance that MYO7A may have on melanocyte homeostasis. Moreover, another MYO7A variant (rs2276288) was associated with increased melanoma susceptibility [30].

In Family 3 we identified two rare variants of WRN. The WRN gene belongs the RecQ subfamily and the DEAH subfamily of DNA and RNA helicases. Consequently, it is associated with DNA transcription, replication, recombination, and repair. Mutants cause the Werner syndrome, an autosomal recessive disorder characterized by progeria and elevated cancer risk. Two variants (rs4733225 and rs13251813) were associated to higher predisposition in melanoma-prone families [31]. Another study of 189 Werner syndrome patients observed that 13.3% developed melanoma, representing a 53-fold elevated risk [32]. Interestingly, one of the variants identified by our study (c.2983G>A; p.Ala995Thr) is contained on the RQC domain, which is responsible for WRN protein-mediated telomere repair [33].

Lastly, we identified a rare variant in the NOP10 gene (NOP10 Ribonucleoprotein), which interacts directly with TERT gene. NOP10 is a member of the telomerase ribonucleoprotein complex that is responsible for telomere maintenance, thus preserving chromosomal integrity and genome stability [34]. Telomere maintenance genes such as TERT, ACD, POT1 and TERF2IP were associated to melanoma predisposition previously [6-8]. The mutant residue that we found (c.34G>C; p.Asp12His) was described previously in a study of congenital dyskeratosis [35].

We have also compared the 288 genes prioritized in our study with candidate genes reported in nine previous genomic studies of hereditary melanoma [14–16, 36–41] (S7 Table) and only one common gene was identified (FANCA). The FANCA gene was identified with a suggestive association to melanoma (p = .002) in the TCGA cohort by Yu et al [37]. FANCA gene DNA repair gene associated with autosomal recessive Fanconi anemia type A, and there is some preliminary evidence of the association of monoallelic pathogenic variants in FANCA and Hereditary Breast and Ovarian Cancer [42] and prostate cancer [43].

We acknowledge that our study has several limitations. First, we only evaluated a small number of families and patients. Second, all of our family duos comprised first degree relatives, which increases the number of shared variants, since any given variant has a 50% chance of being shared between the individuals, do not allow proper linkage analysis and can obscure the identification of pathogenic variants. Third, putative predisposition variants in non-coding or uncaptured regions of the genome (promoter or deep intronic variants) are not detectable by WES.

Also, we did not investigate possible combinatorial effects of more common variants or low penetrance alleles, such as those observed in MC1R genes. The MC1R gene (melanocortin-1 receptor) is one of the main low/moderate penetrance genes related to cutaneous melanoma. MC1R protein regulates the melanogenesis during exposure to UV radiation and, therefore, has a fundamental role in cutaneous pigmentation [44]. The MC1R gene is highly polymorphic, with more than 200 variant alleles been described. Variants called red hair color (RHC) or R alleles are associated with a higher risk (2X) for the development of melanoma as they present loss of receptor function, determining a phenotype of fair skin, ephelides and photosensitivity, in addition to red hair [19, 45, 46]. Non-RHC variants or r alleles determine reduced receptor function and confer less risk (1-2X) for the development of melanoma [19, 45, 47]. In our patients, in two families both relatives harbored MC1R variants previously associated with increased melanoma risk (R and/or r alleles) and one family had one relative with two r alleles. The most frequent variant was the r allele p.Val60Leu, which is associated to a 1.47 [19] increased risk of melanoma and was identified in 3 of 10 patients. An R allele (p.Arg160Trp; associated to a 2.69 [19] increased risk of melanoma) was identified in two patients from distinct families.

Finally, although we cannot conclude that any of the identified variants are the definitive cause of melanoma predisposition in these families, our results represent the first WES data from melanoma-prone families in a highly admixed population and provide a set of rare variants with potential roles in melanoma predisposition. The data from our study can contribute to the future identification of genetic similarities between patients evaluated in different studies, facilitating gene discoveries, and furthering the understanding of molecular mechanisms of melanoma carcinogenesis.

5. Conclusions

Our data revealed rare germline alterations segregating in patients with familial melanoma, providing new knowledge regarding melanoma predisposition in the Brazilian population. By performing in silico analyses of gene function, gene pathways, and variant pathogenicity prediction, we underscored the putative role of particular genes for melanoma risk, contributing with a novel set of potential candidate genes that can be explored further in future studies.

TVC X GATK variant calling.

(XLSX)

Click here for additional data file.

288 variants prioritized variants.

(XLSX)

Click here for additional data file.

Targeted NGS validated variants.

(XLSX)

Click here for additional data file.

VarElect analysis.

(XLSX)

Click here for additional data file.

Probably pathogenic variants (algorithms).

(XLSX)

Click here for additional data file.

IPA main diseases and conditions.

(XLSX)

Click here for additional data file.

Published melanoma candidate genes.

(XLSX)

Click here for additional data file.

20 Apr 2021

PONE-D-21-06922

Family-based whole-exome sequencing identifies rare variants potentially related to cutaneous melanoma predisposition in  Brazilian melanoma-prone families

PLOS ONE

Dear Dr. Carraro,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Both reviewers are experts in the field and agreed that the limited sample size does not fully support the conclusions. Among other issues, the reviewers pointed out that gene candidates may be non-significantly associated after a more rigorous statistical analysis, imposing difficulties in interpreting the enrichment analyses. The reviewers also raised the possibility rare founder variants may be missed by excluding the variants detected in all ten samples. Besides, the limitations in this study may have resulted in identifying variants without a relationship with melanoma susceptibility. On the other hand, the reviewers believe there is merit in this study and found it particularly interesting because reports of genetic variants associated with melanoma are scarce for the Brazilian population. Therefore, I invite the authors to address all concerns and comments below, carefully discussing the limitations of the data and avoiding overstatements.

Please submit your revised manuscript by Jun 04 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see:  http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at  https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Danillo G Augusto

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: A study by Fidalgo, Torrezan and colleagues analyzes rare germline variants in families with enriched cutaneous melanoma history. Unfortunately, the number of samples analyzed is small and significantly affects the power of the study, but authors preformed multiple additional analyses to understand the role of identified putative disease genes.

Several questions that need to be addressed to improve understanding of the study and ability to evaluate the results.

Major:

1. It is great that authors tried to validate NGS data and computational pipelines. Some additional information should be included in the manuscript to fully evaluate the validation results.

1.1. As stated in methods (Lines 89-92), Ion Torrent protocol was validated with GATK best practices pipeline. Could authors include a table/figure showing how concordant the results of variant calling with these pipelines were? What are the number of SNPs and Indels per sample?

1.2. Targeted NGS validation is used to validate original NGS sequencing. Conventionally, a short of variants is validated with qPCR/Sanger sequencing – to change both sequencing technology and computational pipeline. Also, coverage > 10X and VAF>20% are incomplete criteria for validation. Were the genotypes the same? It is not clear from the described approach. If so, could you please state the concordance rate?

2. Comparison with non-cancer exomes, used as one of the filtration steps in variant prioritization should be described in more details. Were the samples jointly called? How was systematic difference between case and control dataset assessed (e.g. using number and frequencies of common synonymous variants?

3. Formal statistical testing for significance of co-segregation is missing. It would be great to have any test (e.g. LOD score) performed to show how significant is co-segregation of identified variants with the phenotype, especially, given a correction for multiple test hypothesis. All the downstream pathway analyses relies on co-segregation which was not formally assessed for significance, therefore, is hard to reliably interpret.

Minor:

link [39] is out of order on line 73.

Reviewer #2: The authors conducted whole exome sequencing on 10 affected individuals from 5 melanoma-prone familiesThe authors conducted whole exome sequencing on 10 affected individuals from 5 melanoma-prone families negative for mutations in CDKN2A and CDK4. The evaluations revealed 288 rare co-segregating coding variants across the 5 families. The authors conducted in silico, gene and pathway-based evaluations to further prioritize variants/genes for further follow-up. Based on these additional evaluations and review of the literature, the authors prioritize several potential candidate genes for further study.

Abstract. Given the small sample size in this study, the authors should modify the conclusions to indicate that the proposed genes are potential candidates. With the current study, it is not possible to conclude which genes are true candidates for melanoma genetic susceptibility.

Materials and Methods. The authors excluded variants if they were detected in all 10 samples. Although such variants would likely be sequencing artifacts or polymorphisms, there is the remote possibility that the authors might have detected a rare founder variant in their population that was responsible for disease in all 5 families. Did the authors evaluate any of the population-level rare variants observed in all 10 samples to make certain that they did not potentially reflect disease-related founder variants?

Depth coverage>50 was used as a filtering criteria. However, the authors report that the sequencing was conducted such that an average of 86% of the target bases were covered more than 20X. What proportion of variants were thus excluded based on depth coverage > 50 being required for retention of variants for further study?

Results. The authors technically validated a subset of the 288 prioritized variants. Were any family members available for further co-segregation evaluation of the prioritized variants in any of the 5 sequenced families?

Discussion. The authors acknowledge the limitations related to the small sample size of their study. Did the authors interrogate publicly available archives/databases to search for data from other melanoma cases and families to further investigate the prioritized variants/genes identified in this study?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

21 Jun 2021

To Prof. Dr. Danillo G Augusto

Academic Editor, PlosOne

We are submitting a revised version of the manuscript “PONE-D-21-06922” for your reconsideration. The manuscript entitled “Family-based whole-exome sequencing identifies rare variants potentially related to cutaneous melanoma predisposition in Brazilian melanoma-prone families” was reviewed according to the suggestions made by the reviewers.

Below we present the point-to-point response to each comment and suggestion. All authors agree with the final version of the manuscript.

I hope you find it appropriate for publishing in PlosOne as a Research Article.

Very best wishes,

Dirce Maria Carraro

Genomics and Molecular Biology Group, International Research Center

A.C. Camargo Cancer Center, R. Taguá, 440, Liberdade, São Paulo – SP, Brazil.

e-mail: dirce.carraro@accamargo.org.br

Reviewer #1: A study by Fidalgo, Torrezan and colleagues analyzes rare germline variants in families with enriched cutaneous melanoma history. Unfortunately, the number of samples analyzed is small and significantly affects the power of the study, but authors preformed multiple additional analyses to understand the role of identified putative disease genes.

Several questions that need to be addressed to improve understanding of the study and ability to evaluate the results.

Response: We thank reviewer one for the careful assessment of our manuscript. We understand the concern of the reviewer regarding the small number of patients included in the study. However, we would like to clarify that by the time of patients’ selection, we have attempt to contact relatives from other melanoma-prone families from our cancer registry (which included 50 individuals with clinical criteria for hereditary melanoma screened for mutations in the main associated genes), and due to the difficulty of including alive affected relatives, we were successful in including only the 10 patients from 5 families that are described in our manuscript. Also, we need to clarify that the funding we had available for performing this project was limited, which was impeditive for performing WES in more individuals, and was the reason why we have opted to focus on a family-based study of highly selected individuals. During the past months, we have tried to recruit other melanoma affected family members for 2 of the 5 families from our study, in other to improve the variant prioritization of these families; however, we were not able to recruit any additional relatives.

Although we understand the limitations that this small number of patients impose to our findings, our results represent the first WES data from melanoma-prone families in Brazil, a highly admixed population with scarce published genomic data. The identified set of rare variants with putative roles in melanoma predisposition can contribute to the future identification of genetic similarities between patients evaluated in different studies, advancing the understanding of molecular mechanisms of melanoma carcinogenesis.

Major:

1. It is great that authors tried to validate NGS data and computational pipelines. Some additional information should be included in the manuscript to fully evaluate the validation results.

Response: We included a more detailed information regarding the validation of prioritized variants, as described in detail in question 1.2.

1.1. As stated in methods (Lines 89-92), Ion Torrent protocol was validated with GATK best practices pipeline. Could authors include a table/figure showing how concordant the results of variant calling with these pipelines were? What are the number of SNPs and Indels per sample?

Response: We have included a supplementary table (S1) containing the total number of variants called by each pipeline, as well as the concordant number of variants and unique number of variants. Briefly, the mean number of SNPs called by TVC was 33,563 and by GATK was 26,747, with a concordance mean of 26,123 SNPs. For indels, the mean number of called variants was 1,654 for TVC and 987 for GATK, with a mean concordance of 741.

1.2. Targeted NGS validation is used to validate original NGS sequencing. Conventionally, a short of variants is validated with qPCR/Sanger sequencing – to change both sequencing technology and computational pipeline. Also, coverage > 10X and VAF>20% are incomplete criteria for validation. Were the genotypes the same? It is not clear from the described approach. If so, could you please state the concordance rate?

Response: We corrected this information in the manuscript and included more detailed data regarding the validation of prioritized variants (Methods section, line 119; Result section line 168). The complete list of validate variants, their coverage, allele frequency and zygosity in both the WES and the targeted NGS sequencing is now presented in the supplementary table S3. We would like to clarify that although the criteria of coverage >10X and VAF >20% were used for setting the variant caller software, our validation data had coverage and frequencies much higher. The mean coverage of the validated variants was 16,217 (range 32 – 147,579) and the mean variant allele frequency was 0.5 (0.42 – 0.61). All variants were confirmed as heterozygous, as expected from WES data. When considering the number of candidate variants that were validated, the correct number is 66 from the 288 candidate variants (in the previous version of the manuscript we mentioned 79 variants, but 13 of these variants were excluded from our final 288 candidate variants list after being validated, due to new prioritization criteria).

2. Comparison with non-cancer exomes, used as one of the filtration steps in variant prioritization should be described in more details. Were the samples jointly called? How was systematic difference between case and control dataset assessed (e.g. using number and frequencies of common synonymous variants?

Response: The non-cancer exome samples were called individually and were provided from another study of one co-author. As in the time of our initial analysis and variant selection (performed in 2016) the ABraOM database was not available (which now contains exome sequencing data for 1,171 Brazilian healthy individuals), these 20 non-cancer individuals were used only to exclude variants that could be common in the Brazilian population and absent in another genomic database. For this filter, we have only received a VCF file containing the variants from the non-cancer samples and compared with variants from our patients. No additional comparison was performed between these cohorts.

3. Formal statistical testing for significance of co-segregation is missing. It would be great to have any test (e.g. LOD score) performed to show how significant is co-segregation of identified variants with the phenotype, especially, given a correction for multiple test hypothesis. All the downstream pathway analyses rely on co-segregation which was not formally assessed for significance, therefore, is hard to reliably interpret.

Response: All our family duos consist of first-degree relatives, which leads to a percentage of shared variants of 50% by chance. To be able to perform a linkage analysis with significance and obtain significant LOD scores several members of the same family, with different degrees of kinship, should be evaluated. This is a recognized limitation of the study and is now better described in the discussion section of the manuscript (line 81): “Second, all of our family duos comprised first degree relatives, which increases the number of shared variants, since any given variant has a 50% chance of being shared between the individuals, do not allow proper linkage analysis and can obscure the identification of pathogenic variants”.

Minor:

link [39] is out of order on line 73.

Response: Literature citations order was corrected in the manuscript.

Reviewer #2: The authors conducted whole exome sequencing on 10 affected individuals from 5 melanoma-prone families. The authors conducted whole exome sequencing on 10 affected individuals from 5 melanoma-prone families negative for mutations in CDKN2A and CDK4. The evaluations revealed 288 rare co-segregating coding variants across the 5 families. The authors conducted in silico, gene and pathway-based evaluations to further prioritize variants/genes for further follow-up. Based on these additional evaluations and review of the literature, the authors prioritize several potential candidate genes for further study.

Response: We thank reviewer two for the careful assessment of our manuscript. We have addressed the reviewer questions below.

Abstract. Given the small sample size in this study, the authors should modify the conclusions to indicate that the proposed genes are potential candidates. With the current study, it is not possible to conclude which genes are true candidates for melanoma genetic susceptibility.

Response: We agree with the reviewer that the identified genes are potential candidates. We have modified the abstract and parts of the discussion and conclusion to avoid overstatement of our findings.

Materials and Methods. The authors excluded variants if they were detected in all 10 samples. Although such variants would likely be sequencing artifacts or polymorphisms, there is the remote possibility that the authors might have detected a rare founder variant in their population that was responsible for disease in all 5 families. Did the authors evaluate any of the population-level rare variants observed in all 10 samples to make certain that they did not potentially reflect disease-related founder variants?

Response: We did not evaluate the presence of any of the population-level rare variants observed in all 10 samples in our first analyses. To answer the request of reviewer 2, we had re-analyzed our data and identified 5,085 variants that were present in all 10 patients. After excluding variants with MAF >0.5% in ExAc and ESP, classified as benign in Clinvar and non-coding variants (except splice site variants), there were 49 variants present in all 10 patients. 39 variants were in 2 genes that presented bad alignment in the regions of the variants. The remaining 10 variants were visually inspected, and 9 were most likely sequencing artifacts while one was most likely a variant present in one of the pseudogenes from CDC27 gene.

Depth coverage>50 was used as a filtering criteria. However, the authors report that the sequencing was conducted such that an average of 86% of the target bases were covered more than 20X. What proportion of variants were thus excluded based on depth coverage > 50 being required for retention of variants for further study?

Response: We apologized for the incorrect value in the first version of the manuscript, we corrected the coverage for 20X (methods: line 106).

Results. The authors technically validated a subset of the 288 prioritized variants. Were any family members available for further co-segregation evaluation of the prioritized variants in any of the 5 sequenced families?

Response: We have tried extensively to recruit other melanoma affected family members for 2 of the 5 families from our study that had more melanoma patients, in other to improve the variant prioritization of these families. However, unfortunately we were not able to recruit any additional affected relatives. We were able to recruit one unaffected member of one family; however, as the underlying genetic causes of melanoma predisposition in these families could be related to genes with moderate or low penetrance, it would be of limited relevance to perform the co-segregation of identified variants in unaffected family members.

Discussion. The authors acknowledge the limitations related to the small sample size of their study. Did the authors interrogate publicly available archives/databases to search for data from other melanoma cases and families to further investigate the prioritized variants/genes identified in this study?

Response: We did not analyze raw data from other melanoma studies. However, we performed an extensive literature review to identify other studies of that investigated candidate genes to melanoma predisposition. We have identified 9 studies that used comprehensive genomic approaches (WES and WGS) in melanoma patients and compared the candidate genes found in these studies (supplementary table S7) with the list of genes prioritized in our study. From our 288 initial candidates, only one the FANCA was identified as a candidate gene to melanoma risk in other study (Yu et al, 2018). A paragraph regarding these comparisons was added to the discission section (lines 73-78).

Submitted filename: 2021.06.18 - Response to Reviewers.pdf

Click here for additional data file.

11 Jul 2021

PONE-D-21-06922R1

Family-based whole-exome sequencing identifies rare variants potentially related to cutaneous melanoma predisposition in  Brazilian melanoma-prone families

PLOS ONE

Dear Dr. Carraro,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands.

Some of the issues raised by the reviewers were not fully addressed. Therefore, we invite you to submit a revised version of the manuscript that addresses the remaining points.

Please submit your revised manuscript by Aug 25 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see:  http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at  https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .

We look forward to receiving your revised manuscript.

Kind regards,

Danillo G Augusto

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: N/A

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The updated manuscript by Fidalgo, Torrezan and colleagues has been significantly improved. Clarification of the technical processing of the data aid the interpretability of the study.

For the comparison with non-cancer exomes I would still insist on performing a technical calibration of the sequencing data. Since case and control data was not jointly called technical differences between the variant calling pipelines could significantly bias the comparison. To prove the validity of this comparison, I would like to see the comparison of the synonymous variants between cases and controls. As you mentioned in your response letter, you have got the VCF file from controls with all variants. Allele frequencies comparison between cases and controls on synonymous variants could be performed with already available data relatively quickly. QQ-plot could be generated to evaluate the distribution of the test statistic.

Reviewer #2: The authors attempted to address my comments. They were unable to recruit additional family members and so were not able to collect additional information on the 5 families evaluated.

There is an error in line 92 of the revision. Part of the sentence appears to be missing.

The authors should check the manuscript for grammar errors.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

14 Sep 2021

To Prof. Dr. Danillo G Augusto

Academic Editor, PlosOne

We are submitting a revised version of the manuscript “PONE-D-21-06922” for your reconsideration. The manuscript entitled “Family-based whole-exome sequencing identifies rare variants potentially related to cutaneous melanoma predisposition in Brazilian melanoma-prone families” was reviewed according to the remaining suggestions made by the reviewers.

Below we present the point-to-point response to each comment and suggestion. All authors agree with the final version of the manuscript.

I hope you find it appropriate for publishing in PlosOne as a Research Article.

Very best wishes,

Dirce Maria Carraro

Genomics and Molecular Biology Group, International Research Center

A.C. Camargo Cancer Center, R. Taguá, 440, Liberdade, São Paulo – SP, Brazil.

e-mail: dirce.carraro@accamargo.org.br

Reviewer #1: The updated manuscript by Fidalgo, Torrezan and colleagues has been significantly improved. Clarification of the technical processing of the data aid the interpretability of the study.

For the comparison with non-cancer exomes I would still insist on performing a technical calibration of the sequencing data. Since case and control data was not jointly called technical differences between the variant calling pipelines could significantly bias the comparison. To prove the validity of this comparison, I would like to see the comparison of the synonymous variants between cases and controls. As you mentioned in your response letter, you have got the VCF file from controls with all variants. Allele frequencies comparison between cases and controls on synonymous variants could be performed with already available data relatively quickly. QQ-plot could be generated to evaluate the distribution of the test statistic.

Response: We thank reviewer one for the first and second assessment of our manuscript and the important suggestions and commentaries. We were able to perform the comparison of the allele frequencies of synonymous variants between cases and controls. As it is can be seen in the graphics submitted in the PDF rebuttal letter (A: Density; B: Empirical cumulative distribution function), there is only a minor difference observed between cases and controls, which can be due to the different sequencing methods applied, different variant calling pipelines and different coverage mean, which was higher in controls than in patients. However, we believe this minor differences do not impact the purpose of using these non-cancer exome samples, as they were used only to exclude variants that could be common in the Brazilian population and absent in another genomic database, since at that time of our first data analysis and prioritization (performed in 2016) the ABraOM database was not available (database with exome sequencing data for 1,171 Brazilian healthy individuals, that was included in a second filter of our analysis).

Reviewer #2: The authors attempted to address my comments. They were unable to recruit additional family members and so were not able to collect additional information on the 5 families evaluated.

There is an error in line 92 of the revision. Part of the sentence appears to be missing.

The authors should check the manuscript for grammar errors.

Response: We thank reviewer one for the first and second assessment of our manuscript and the important suggestions and commentaries.

We apologized for the missing sentence. It was corrected to “The comparison between the number of variants called by each pipeline and the numbers of concordant calls are described in the supplementary table (S1).” (methods: line 92).

We have performed a detailed revision of the manuscript to correct grammar errors.

Submitted filename: 2021.08.14 - Rebuttal letter.pdf

Click here for additional data file.

13 Oct 2021

13 Dec 2021

To Prof. Dr. Danillo G Augusto

Academic Editor, PlosOne

We are submitting the rebuttal letter for the third review of the manuscript “PONE-D-21-06922” entitled “Family-based whole-exome sequencing identifies rare variants potentially related to cutaneous melanoma predisposition in Brazilian melanoma-prone families” for your reconsideration.

First, we would like to apologize for the delay in submitting our response. Since the reviewer requested a novel analysis for which none of the authors of this study had sufficient expertise, we had to consult other experts in the field of population genetics.

Below we present the response to the comments of the reviewers. In previous revisions, the manuscript was reviewed according to all the suggestions made the reviewers. In this version, as justified below in our letter, we have only revised the reference list to ensure that it is complete and correct, as requested by the Journal. All authors agree with the final version of the manuscript.

I hope you find it appropriate for publishing in PlosOne as a Research Article.

Very best wishes,

Dirce Maria Carraro

Genomics and Molecular Biology Group, International Research Center

A.C. Camargo Cancer Center, R. Taguá, 440, Liberdade, São Paulo – SP, Brazil.

e-mail: dirce.carraro@accamargo.org.br

Reviewer #1: The manuscript coherently outlines findings and analyses. However, I am still not convinced by the figures presented in the rebuttal letter. Since authors have allele frequencies in hand - there is a standard way of evaluating the technical bias in the data. Performing association study on synonymous variants and generating QQ-plot using obtained p-values. Genomic inflation (lambda) will become a numeric criterion for "matching" of the case and control data. I would like to kindly request this figure to be added to the supplement. Comments on the genomic inflation estimate (inflated/non-inflated) should be added to the main text.

Response: We thank reviewer one for the assessment of our manuscript. We understand the remaining concerns of the reviewer regarding the control group and the request for performing genomic inflation estimate. However, we would respectfully like to argue that, unlike GWAS studies and other types of case-control studies that compare large populations and select candidate variants through a comparison of frequencies between cases and controls, our study has a different design. Our study is a family-based study, with the goal of identifying very rare genetic variants that are segregating in affected members of five families with high melanoma risk. In this scenario, the small number of control samples (10 Brazilian individuals) were used only to exclude variants that could be unique in the Brazilian population and absent in another genomic database, since at that time of our first data analysis and prioritization (performed in 2016) the ABraOM database was not available (database with exome sequencing data for 1,171 Brazilian healthy individuals, that was included in a second filter of our analysis). In this sense, it is of our understanding that since we did not used the controls to perform association analysis, given the small number of both controls (10 individuals) and patients (10 from 5 families), and the kinship relationship of the cases, comparing these groups with Q-Q plot analysis would not be appropriated.

Reviewer #2: All comments have been addressed.

Response: We thank reviewer one for the assessment of our manuscript and the important suggestions and commentaries in the previous rounds of revision.

Submitted filename: 2021.12.07 - Response to reviewers.pdf

Click here for additional data file.

26 Dec 2021

Family-based whole-exome sequencing identifies rare variants potentially related to cutaneous melanoma predisposition in  Brazilian melanoma-prone families

PONE-D-21-06922R3

Dear Dr. Carraro,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Danillo G Augusto

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I am respectfully accepting your argument, though, as a suggestion, I would recommend that the brief statement about inability/irrelevance of genetic background matching is added to the manuscript methods/discussion section.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

17 Jan 2022

PONE-D-21-06922R3

Family-based whole-exome sequencing identifies rare variants potentially related to cutaneous melanoma predisposition in  Brazilian melanoma-prone families

Dear Dr. Carraro:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Danillo G Augusto

Academic Editor

PLOS ONE