Novel APC promoter and exon 1B deletion and allelic silencing in three mutation-negative classic familial adenomatous polyposis families
Genome Medicinevolume 7, Article number: 42 (2015)
The overwhelming majority (approximately 80%) of individuals with classic familial adenomatous polyposis (FAP) exhibit mutations in the coding sequence of the adenomatous polyposis coli (APC) tumor suppressor gene. Families without detectable APC mutations are unable to benefit from the use of genetic testing for clinical management of this autosomal dominant syndrome.
We used exome sequencing and linkage analysis, coupled with second-generation sequencing of the APC locus including non-coding regions to investigate three APC mutation-negative classical FAP families.
We identified a novel ~11 kb deletion localized 44 kb upstream of the transcription start site of APC that encompasses the APC 1B promoter and exon. This deletion was present only in affected family members of one kindred with classical FAP. Furthermore, this same deletion with identical breakpoints was found in the probands of two additional APC mutation-negative classical FAP kindreds. Phasing analysis of single nucleotide polymorphisms (SNPs) around the deletion site in the three probands showed evidence of a shared haplotype, suggesting a common founder deletion in the three kindreds. SNP analysis within the coding sequence of APC, revealed that this ~11 kb deletion was accompanied by silencing of one of the APC alleles in blood-derived RNA of affected individuals.
These results support the causal role of a novel promoter deletion in FAP and suggest that non-coding deletions, identifiable using second-generation sequencing methods, may account for a significant fraction of APC mutation-negative classical FAP families.
Familial adenomatous polyposis (FAP) is an inherited, autosomal dominant syndrome characterized by the development of colonic polyposis and cancer progression by 40 years of age when left untreated. The classical FAP phenotype results in the growth of hundreds to thousands of adenomas, caused in approximately 70% to 80% of affected families by germline mutations in the tumor suppressor gene adenomatous polyposis coli (APC) on chromosome 5q21-22 [1,2]. An attenuated FAP phenotype results in generally fewer than 100 adenomas, and fewer than 10% of such families have detectable APC mutations . Mutations in the gene Mut Y homolog (MUTYH) have been associated with an autosomal recessive form of inherited colonic polyposis and are not observed in classical or attenuated FAP .
Hundreds of pathologic mutations in APC have been described which result in FAP, most by producing a truncated protein product . Such mutations occur throughout the 15 exons of APC, although mutation hot spots occur between codons 1,000 to 1,500 of exon 15 [2,3]. Routine clinical genetic testing has therefore consisted of Sanger sequencing of the 15 exons of APC as well as southern blot analysis on all 15 exons to detect rearrangements or copy number variations . The inability to determine a causal mutation of FAP in ‘APC mutation-negative’ families precludes the identification of unaffected members and necessitates intensive colonic surveillance of all individuals of the family .
Reports have begun to describe mutations surrounding exon 1B of APC in mutation-negative families: A ~61 kb deletion which partially overlapped promoter 1B was described in a Swedish FAP family . A large deletion containing promoter 1A was found in a Mennonite family through multiplex ligation-dependent probe amplification, but the genomic breakpoint coordinates were not defined . A ~21 kb deletion containing promoter 1B was described in the proband and affected father of a Bulgarian family . More recently, distinct promoter deletions were identified in APC in mutation-negative families which were approximately 34 kb  and approximately 132 kb  in size. These deletions are particularly relevant given prior observations that one APC allele is silenced in many individuals with APC mutation-negative FAP .
Here we investigated a multi-generational APC mutation-negative FAP kindred, using second-generation sequencing techniques that identified an ~11 kb deletion which encompasses exon 1B and the associated promoter . Furthermore, we found this deletion in two additional FAP kindreds previously classified as APC mutation-negative by standard genetic testing. In the probands of all three kindreds, we demonstrate the absence of a transcript from one APC allele indicating that the ~11 kb deletion likely silences APC expression. These results highlight the importance of second-generation sequencing approaches to test for potentially pathologic mutations of APC which cannot be detected by current standard genetic tests.
Within our FAP registry, we identified three APC mutation-negative FAP kindreds with classical phenotypes. Participants from these kindreds gave written informed consent for participation in a study approved by the Institutional Review Board of Washington University, St. Louis (IRB# 201105464) and which conforms to the Helsinki Declaration. These kindreds were concentrated in three different states: Missouri, Illinois, and Idaho, and were not known to be related. The probands from each kindred underwent clinical genetic testing with Colaris AP® (Myriad Genetic Laboratories, Salt Lake City, UT, USA), which sequences 15 exons of APC and exons 7 and 13 of MUTYH to detect the mutations Y165C and G382D. Southern blot analysis on all exons of APC is also performed to detect large rearrangements. No known pathologic mutations in APC and MUTYH were detected by this method in the probands of the three kindreds.
Blood from participants was collected in tubes without additives for DNA extraction and in PAXgene Blood RNA Tubes (PreAnalytiX; Qiagen, Valencia, CA, USA) for RNA preservation and extraction. DNA blood extraction was performed using the Gentra Puregene Blood Kit (Qiagen), and RNA blood extraction was performed using the PAXgene Blood RNA Kit (Qiagen).
Exome sequencing and linkage analysis
Exome capture was performed with the SureSelect 38 Mb All Exon kit (Agilent, Salta Clara, CA, USA) for members of kindred 0124 III-5, III-3, III-4, and the SureSelect 72 Mb (V4 + UTR) kit was used for individuals 0124 IV-4, V-6, IV-2, V-1, and IV-3. Paired-end sequencing (2 × 101 bp) was performed on the HiSeq 2000 platform (Illumina, San Diego, CA, USA). The sequences were aligned against the GRCh37/hg19 human reference genome using Novoalign (Novocraft, Selangor, Malaysia), optical duplicate sequence removal was performed using samtools , and variant calls were identified using samtools and GATK . Local realignment and base quality recalibration using GATK were performed prior to variant calling, and a genotype quality GQ threshold of 20 was used to retain single nucleotide variants (SNVs) for further analysis. All reported genomic coordinates are in relation to the GRCh37/hg19 reference human genome. The sequencing data is available at dbGaP with accession number phs000904.v1.p1.
To search for candidate exon mutations which segregated with the FAP phenotype in kindred 0124, we applied a series of filters to the identified SNVs. We required that a candidate SNV follow an autosomal pattern of inheritance and retained only those which were heterozygous in affected individuals and homozygous in unaffected ones. The SNVs were annotated for function using ANNOVAR , and only non-synonymous mutations were retained. Finally, single nucleotide polymorphisms (SNPs) in dbSNP132  and identified in the 1000 Genomes project  were removed. To perform linkage analysis, the unfiltered, aggregated SNVs were analyzed with MERLIN using parametric linkage analysis with a rare dominant disease model .
Coloseq™ testing and deletion validation
Subjects IV-4 of kindred 0124, IV-1 of kindred 0130, and III-2 of kindred 0163 underwent sequencing of the APC locus with the ColoSeq™ Polyposis Panel, which performs second-generation sequencing of APC and MUTYH to cover all exons, introns and flanking sequences . An ~11 kb deletion encompassing exon and promoter 1B was detected in all three individuals between coordinates chr5:112,034,824-112,045,845. To verify the presence of this deletion, PCR using standard conditions and primers P1, P2, and P3 (Table 1) were used. In the presence of template genomic DNA with the ~11 kb deletion, primers P1 and P3 form a 298 bp PCR product, whereas primers P1 and P2 do not result in a product because the binding site for primer P2 is absent. With a normal genomic DNA template, primers P1 and P2 form a 373 bp product, whereas primers P1 and P3 do not form a product because the distance of their binding sites exceeds 11 kb. The PCR amplicons were sequenced with Sanger chemistry using the ABI BigDye Terminator Mix (version 3, Applied Biosystems, Foster City, CA, USA).
SNP genotyping and haplotype phasing
To query SNPs flanking the promoter deletion, the following primer pairs were used (Table 1): primers P4/P5 for rs76768628, rs60905866, and rs10463643; primers P6/P7 for rs10051624; primers P8/P9 for rs2900066, rs6594643, and rs6594644; primers P10/P11 for rs4705559 and rs7704618; and primers P12/P13 for rs1661035 and rs76028897. To amplify DNA specifically from the chromosomal strand harboring the ~11 kb deletion, two primer pairs were used: P8/P13 and P14/P15 (Table 1). On chromosomes with the ~11 kb deletion, these primer pairs resulted in amplicons of 5,541 bp and 6,956 bp in size, respectively. Because these amplicons represent DNA amplified only from the chromosomal strand with the ~11 kb deletion, all of the SNP genotypes determined from these amplicons were used as templates for Sanger sequencing (and haplotype phasing) using primers P4, P6, P8, P10, and P12 (Table 1).
Allelic expression assessment
dbSNP was searched for SNPs residing in the coding region of APC, and rs459552 and rs465899 were found to be heterozygous in the exome sequence data of one or more individuals of kindred 0124. After extraction of RNA from blood collected in PAXgene Blood RNA Tubes, reverse transcription was performed using SuperScript III Reverse Transcriptase (Invitrogen, Grand Island, NY, USA). The following primers were used to perform PCR and subsequent sequencing for SNP assays (Table 1): primers P16/P17 for rs459552 (producing amplicon chr5:112,176,294-112,176,979) and primers P18/P19 for rs465899 (producing amplicon chr5:112,176,956-112,177,213).
Exome sequencing of a classical FAP family reveals a LOD peak at the APC locus
Kindred 0124 is a multi-generational family with a classical FAP phenotype (Figure 1A), including the presence of multiple extra-colonic malignancies but with no mutations in APC using standard genetic testing. To investigate the causal mutation accounting for FAP, we performed exome sequencing on eight individuals spanning three generations of the 0124 kindred. Analysis of the exome data revealed one non-synonymous heterozygous mutation which segregated in the affected family members and did not appear in dbSNP132 or an in-house exome sequence database. This was a c.G512A (p.R171Q) mutation in the gene Proline Rich Protein BstNI Subfamily 2 (PRB2) at coordinate chr12:11,546,500. PRB2 is a member of a class of genes located on chromosome 12p13.2 which code for abundant proteins in salivary excretion . Of note, this SNP has appeared in dbSNP as of version 135 and has a minor allele frequency of 0.009 in dbSNP 137. The significance of this non-synonymous mutation remains unclear in relation to the FAP phenotype.
Considering the possibility of a non-coding mutation accounting for the FAP phenotype, we turned to linkage analysis using SNPs derived from exome sequencing data. Using parametric linkage analysis with a dominant disease model , we found several loci with positive logarithm of the odds (LOD) scores, although none had a LOD score over 3 (Figure 1B). We did not find a positive LOD score around the PRB2 locus, further indicating a lack of support for the role of PRB2 in the FAP phenotype. However, we found two loci with a maximum LOD score of 2.408 - one at 5q22.2 (111.5 to 112.9 Mb) and another at 7p14.1 (37.2 to 43.3 Mb). 5q22.2 notably contains the APC gene and suggested the possibility of a non-coding APC mutation being causal in this FAP kindred. Because of the known relationships between APC and FAP, we chose to focus the remainder of this study at the APC locus.
Second-generation sequencing of the APC locus reveals an 11 kb promoter deletion
To search for a non-coding mutation at the APC locus of the affected individual IV-4 of kindred 0124, we used the ColoSeq™ assay to capture and sequence exonic, intronic, and promoter regions of the APC gene . This assay identified an ~11 kb heterozygous deletion 44 kb upstream of the first exon of APC between coordinates chr5:112,034,824-112,045,845 (Figure 2). This deletion contains exon and promoter 1B (Genbank D13981.1) and represents the smallest promoter deletion currently described in an FAP family.
We verified the presence of the deletion using primers designed to create a PCR product of 373 bp when a genomic template with normal APC promoter is present (primers 1 and 2, Figure 3A). A second primer pair creates a 298 bp PCR product only when a genomic template with the ~11 kb deletion is present (primers 1 and 3, Figure 3A). We used these primers to assess the eight members of kindred 0124 (Figure 3B). PCR in all five affected individuals generated both 298 bp and 373 bp products, indicating heterozygosity for the promoter deletion allele in APC. In the three unaffected individuals, only the 373 bp fragment was generated indicating the absence of the promoter deletion.
Further verifying the presence of the ~11 kb promoter deletion, we performed Sanger sequencing of the PCR products from the proband of kindred 0124. Using primer 3 for Sanger sequencing, we confirmed an 11,020 bp gap in the mapped coordinates from the sequencing trace (Figure 3C). Sanger sequencing using primer 3 also confirmed the presence of a normal allele in the proband (Figure 3D).
11 kb promoter deletion is found in two additional FAP kindreds
Two additional kindreds in our FAP registry were found to be APC mutation-negative by standard genetic testing. These kindreds, 0130 and 0163 (Figure 4A and C), reside in different states in America and have no known familial relationship to each other or to the index kindred 0124. To assess the presence of promoter deletions in these two additional FAP kindreds, we used the ColoSeq™ assay to sequence the APC locus in each proband and surprisingly found the same ~11 kb deletion. Further investigation using Sanger sequencing confirmed the identical deletion coordinates in both probands (Figure 4B and D). To exclude possible sample error, we verified those findings in a second blood sample from the probands of all three kindreds 1 year after obtaining the first samples, reinforcing our conclusions that affected members from these independent kindreds indeed harbor the identical mutation.
Genetic relationship between kindreds revealed by haplotype phasing
As alluded to above, these three kindreds reside in geographically distinct regions and no known relatedness was discerned by questionnaire. Nevertheless, we examined genetic evidence of a relationship between these kindreds by determining the genetic haplotype associated with the ~11 kb deletion in each of the three probands. We first queried 11 SNPs flanking the deletion coordinates (Table 2), seven of which ranged from 229 to 2,408 bp 5′ of the deletion, and four of which ranged from 4,276 to 6,151 bp 3′ of the deletion. SNPs rs10463643, rs10051624, rs2900066, rs6594643, rs6594644, rs4705559, rs7704618, and rs1661035 were heterozygous in the probands of the 0124 and 0163 kindreds. These findings suggested that these SNPs could distinguish the haplotype associated with the ~11 kb deletion from the haplotype from the chromosomal strand without the deletion. Accordingly we phased the SNP genotypes by using a PCR strategy to selectively amplify the DNA from the chromosomal strand containing the ~11 kb deletion (Figure 5). Flanking primers were designed such that amplicons of 5.5 kb and 6.9 kb would reflect the deletion, whereas amplification from the wild type chromosomal strand would yield an amplicon 11 kb larger. Accordingly, we then isolated the 5.5 and 6.9 kb amplicons for Sanger sequencing as a result of which we were able to phase the SNP genotypes associated with the deletion. The sequencing results showed that the haplotype associated with the ~11 kb deletion was identical across the queried SNPs in each of the three probands (Table 2), providing evidence to suggest that these three kindreds share a common ancestor.
APC promoter deletion results in allelic silencing in blood
APC is transcribed in a wide variety of tissues, including peripheral blood cells . To assess the effect of the ~11 kb deletion on allelic transcription in these kindreds, we identified exonic SNPs in APC which were heterozygous in the germline of affected and unaffected individuals of these FAP kindreds. Silencing of an APC allele in individuals heterozygous for these SNPs would result in the detection of only one base at the corresponding complementary DNA (cDNA) site from expression of a single APC allele. Sanger sequencing of genomic DNA (gDNA) from proband III-5 of kindred 0124 showed heterozygousity at SNPs rs459552 and rs465899, but sequencing of cDNA derived from blood shows the presence of only one allele for both corresponding positions. This is in contrast to the unaffected individual III-4 of kindred 0124 whose gDNA is heterozygous for SNP rs465899, and in whom cDNA sequencing shows evidence of mRNA expression from two alleles (Table 3 and Figure 6). The silencing of one APC allele was also observed in affecteds from the other two kindreds with loss of cDNA heterozygosity at SNP rs459552 in proband IV-1 of kindred 0130, and SNPs rs459552 and rs465899 in proband III-2 of kindred 0163 (Table 3 and Figure 6).
Here we report the presence of a novel ~11 kb heterozygous deletion in the APC promoter of three independent APC mutation-negative classical FAP kindreds. Our findings add to the growing body of literature suggesting that non-coding APC deletions may contribute to an important fraction of APC mutation-negative classical FAP. The current literature is based in reports of a founder mutation in a Mennonite kindred, a kindred in the Swiss Polyposis Registry, and a single Bulgarian kindred [6-8]. However, this is the first report to our knowledge that the identical mutation exists in three classical FAP kindreds. In addition, there are informative features of the approach we used to identify this mutation that bear further discussion.
Our evaluation began with the exome sequencing of one multigenerational family with classical FAP to address the possibility that the causal mutation was a coding mutation in a novel gene. We identified a non-synonymous heterozygous mutation that segregated with the polyposis phenotype across the kindred. However, this mutation in PRB2 remains of unclear significance since database searches revealed no plausible connection to FAP or gastrointestinal polyposis in general. Our experience parallels that of others showing some of the potential pitfalls in using exome sequencing to search for causal mutations in Mendelian diseases. For example, in the search for the causal mutation for Kabuki syndrome, Ng et al. performed exome sequencing on 10 unrelated affected individuals and found one rare, non-synonymous mutation (in MUC16) across all 10 individuals . However, this particular mutation was felt to be a false positive result, and further investigations identified a more likely mutation candidate (MLL2).
Although exome sequencing has been used successfully to identify over 100 Mendelian disease genes [21,22], the number of unsuccessful applications of this technique is unknown. Several reasons may result in the inability to identify a mutation using exome sequencing: False negatives can result from technical issues such as poor capture and subsequent low sequence coverage of exons, or from analytical errors such as from indiscriminant candidate filtering using SNPs from public databases. Although the Agilent SureSelect Human All Exon V4 + UTR probe set captured DNA from exon 1B in our study, data from targeted capture and exome sequencing are generally not amenable to the discovery of copy number variations because of technical biases. Sequence coverage of exon 1B did not differ between affected and non-affected members of kindred 0124 (data not shown) and therefore did not reflect the ~11 kb deletion. Although the exome sequencing strategy did not yield directly a causal variant, we were able to perform linkage analysis using SNPs derived from the exome sequences. The linkage data did not yield a LOD score significant at the whole genome level, but of the positive signals identified, the one of obvious interest was at the APC locus.
We further investigated the APC locus, using a second-generation sequencing assay across the non-coding regions to reveal an ~11 kb deletion. Unlike array-based or multiplex ligation-dependent probe amplification (MLPA) techniques used to search for regions of copy number variation, the second-generation sequencing assay was able to identify the exact breakpoint coordinates. These were further confirmed by an orthogonal sequencing modality with Sanger chemistry. Presumably, another benefit of the second-generation sequencing is the ability of the technique to detect diminutive copy number variants down to insertion/deletions of a few bases in length, whereas array or MLPA techniques are able to detect only large variants greater than kilobases in length. Unexpectedly, we found the same deletion breakpoints in all three independent APC mutation-negative kindreds in our FAP registry. The phased typing of SNPs flanking the ~11 kb deletion further showed that the three kindreds shared the same haplotype associated with the deletion, which suggests that the deletion in these kindreds arose from a common ancestor.
This ~11 kb deletion encompasses exon and promoter 1B, and using SNP analysis on APC transcripts, we found that this deletion silenced one APC allele in affected individuals. Rolin et al. described a ~61 kb heterozygous deletion which resulted in a reduction in expression, but not silencing, of an APC allele . This ~61 kb deletion removes a portion of the exon 1B 5′ untranslated region but does affect an open reading frame within it (RefSeq NM_001127511.2). The ~11 kb deletion described in this study, by contrast, deletes exon1B and its upstream sequences in its entirety. It may be that the deletion of the exon 1B 5′ untranslated region and/or its open reading frame results in its targeted degradation. Interestingly however, a ~21 kb deletion which encompasses the ~11 kb deletion described in this study was reported to reduce, but not to silence, allelic expression of APC . The interaction of 1A and 1B promoters in different tissue types and other epigenetic regulatory mechanisms need further investigation. For example, although our results show that the ~11 kb deletion results in diminished transcription of APC in blood, differential promoter use may still result in APC transcript production using promoter 1A in other tissues.
The allelic silencing of APC in mutation-negative kindreds is consistent with observations made over a decade ago. Using a method of isolating individual chromosomes from affected individuals in hybrid rodent cells to perform monoallelic mutation analysis, one APC allele was found to be silenced in seven of nine FAP subjects . The mechanism of allelic silencing could not be determined at that time, however.
We provide evidence that an ~11 kb deletion encompassing exon and promoter 1B of APC which segregates with the FAP phenotype in three affected kindreds results in allelic silencing. Second-generation sequencing techniques is ideal for the detection of such deletions, as they can vary greatly by size and position (Figure 2). Furthermore, the availability of such assays in the clinical setting such as with ColoSeq™ has the potential to readily impact the clinical management of FAP kindreds.
Adenomatous polyposis coli
Familial adenomatous polyposis
Logarithm of the odds
Friedl W, Caspari R, Sengteller M, Uhlhaas S, Lamberti C, Jungck M, et al. Can APC mutation analysis contribute to therapeutic decisions in familial adenomatous polyposis? Experience from 680 FAP families. Gut. 2001;48:515–21.
Moisio A, Järvinen H, Peltomäki P. Genetic and clinical characterisation of familial adenomatous polyposis: a population based study. Gut. 2002;50:845–50.
Sieber OM, Lipton L, Crabtree M, Heinimann K, Fidalgo P, Phillips RKS, et al. Multiple colorectal adenomas, classic adenomatous polyposis, and germ-line mutations in MYH. N Engl J Med. 2003;348:791–9.
Sieber OM, Tomlinson IP, Lamlum H. The adenomatous polyposis coli (APC) tumour suppressor–genetics, function and disease. Mol Med Today. 2000;6:462–9.
Burt R, Neklason DW. Genetic testing for inherited colon cancer. Gastroenterology. 2005;128:1696–716.
Rohlin A, Engwall Y, Fritzell K, Göransson K, Bergsten A, Einbeigi Z, et al. Inactivation of promoter 1B of APC causes partial gene silencing: evidence for a significant role of the promoter in regulation and causative of familial adenomatous polyposis. Oncogene. 2011;30:4977–89.
Charames GS, Ramyar L, Mitri A, Berk T, Cheng H, Jung J, et al. A large novel deletion in the APC promoter region causes gene silencing and leads to classical familial adenomatous polyposis in a Manitoba Mennonite kindred. Hum Genet. 2008;124:535–41.
Kadiyska T, Todorov T, Bichev S, Vazharova R, Nossikoff A, Savov A, et al. APC promoter 1B deletion in familial polyposis-implications for mutation-negative families. Clin Genet. 2014;85:452–7.
Snow AK, Tuohy TMF, Sargent NR, Smith LJ, Burt RW, Neklason DW. APC promoter 1B deletion in seven American families with familial adenomatous polyposis. Clin Genet. 2014 Sep 19. doi:10.1111/cge.12503 [Epub ahead of print].
Pavicic W, Nieminen TT, Gylling A, Pursiheimo JP, Laiho A, Gyenesei A, et al. Promoter-specific alterations of APC are a rare cause for mutation-negative familial adenomatous polyposis. Genes Chromosom Cancer. 2014;53:857–64.
Laken SJ, Papadopoulos N, Petersen GM, Gruber SB, Hamilton SR, Giardiello FM, et al. Analysis of masked mutations in familial adenomatous polyposis. Proc Natl Acad Sci U S A. 1999;96:2322–6.
Pritchard CC, Smith C, Salipante SJ, Lee MK, Thornton AM, Nord AS, et al. ColoSeq provides comprehensive lynch and polyposis syndrome mutational analysis using massively parallel sequencing. J Mol Diagn. 2012;14:357–66.
Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The sequence alignment/map format and SAMtools. Bioinformatics. 2009;25:2078–9.
McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20:1297–303.
Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res. 2010;38:e164.
Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM, et al. dbSNP: the NCBI database of genetic variation. Nucleic Acids Res. 2001;29:308–11.
Abecasis GR, Auton A, Brooks LD, DePristo MA, Durbin RM, Handsaker RE, et al. An integrated map of genetic variation from 1,092 human genomes. Nature. 2012;491:56–65.
Abecasis GR, Cherny SS, Cookson WO, Cardon LR. Merlin–rapid analysis of dense genetic maps using sparse gene flow trees. Nat Genet. 2002;30:97–101.
Azen E, Lyons KM, McGonigal T, Barrett NL, Clements LS, Maeda N, et al. Clones from the human gene complex coding for salivary proline-rich proteins. Proc Natl Acad Sci U S A. 1984;81:5561–5.
Ng SB, Bigham AW, Buckingham KJ, Hannibal MC, McMillin MJ, Gildersleeve HI, et al. Exome sequencing identifies MLL2 mutations as a cause of Kabuki syndrome. Nat Genet. 2010;42:790–3.
Rabbani B, Mahdieh N, Hosomichi K, Nakaoka H, Inoue I. Next-generation sequencing: impact of exome sequencing in characterizing Mendelian disorders. J Hum Genet. 2012;57:621–32.
Gilissen C, Hoischen A, Brunner HG, Veltman JA. Disease gene identification strategies for exome sequencing. Eur J Hum Genet. 2012;20:490–7.
This work was supported by a Pilot and Feasibility award (YL) from the Washington University Digestive Diseases Research Core Center (DDRCC) grant #5P30 DK052574. SL received support from the NIH Fellowship Grants F32HL110473 and K99HL119617. NOD is supported through grants from the NIH (DK-56260, HL-38180, DK-52574) and support from the Buehrle Family fund and the Gale Family fund. We thank the Alvin J. Siteman Cancer Center at Washington University School of Medicine and Barnes-Jewish Hospital in St. Louis, MO, for the use of the Hereditary Cancer Core. The Siteman Cancer Center is supported in part by NCI Cancer Center Support Grant P30 CA91842.
The authors declare that they have no competing interests.
Concepts and design: YL, SL, MDB, PJG, NOD. Collection and assembly of data: All authors. Data analysis and interpretation: YL, SL, PJG, WC, NOD. Manuscript writing: All authors. Final approval of manuscripts: All authors.