Fine-mapping he 2q37 and 17q11.2-q22 loci o no el genes
and sequence a ian s associa ed wi h a gene ic p edisposi ion
o p os a e cance
Vi pi H. Lai inen
1
, Tommi Ran ape o
1
, Daniel Fische
2
, Elisa M. Vuo inen
1
, Teu o L.J. Tammela
3
, PRACTICAL Conso ium,
Tiina Wahl o s
1
and Johanna Schleu ke
1,4
1
BioMediTech, Uni e si y o Tampe e and Fimlab Labo a o ies, FI-33520, Tampe e, Finland
2
School o Heal h Sciences, Uni e si y o Tampe e, FI-33014 Tampe e, Finland
3
Depa men o U ology, Tampe e Uni e si y Hospi al and Medical School, Uni e si y o Tampe e, FI-33520 Tampe e, Finland
4
Medical Biochemis y and Gene ics, Ins i u e o Biomedicine, Uni e si y o Tu ku, FI-20014 Tu ku, Finland
Key wo ds: p os a e cance isk, gene ic p edisposi ion, suscep ibili y loci, 2q37, 17q11.2-q22
Abb e ia ions: AR: And ogen Recep o ; ChIP-seq: Ch oma in Immunop ecipi a ion Combined wi h Massi ely Pa allel DNA Sequencing;
CI: Con idence In e al; DB: Da abase; DE: Di e en ially Exp essed (gene); eQTL: Exp ession Quan i a i e T ai Locus; GWAS: Genome
Wide Associa ion S udy; HPC: He edi a y P os a e Cance ; HWE: Ha dy-Weinbe g Equilib ium; Indel: Inse ion/Dele ion Polymo phism;
LD: Linkage Disequilib ium; LincRNA: La ge In e genic Non-Coding RNA; LNCaP: And ogen-Sensi i e Human P os a e Adenoca ci-
noma Cell Line De i ed F om Lymph Node Me as asis; MAF: Mino Allele F equency; OR: Odds Ra io; P Ca: P os a e Cance ; PSA:
P os a e Speci ic An igen; PWM: Posi ion Weigh Ma ix; RNA-seq: Massi ely Pa allel RNA Sequencing; SNP: Single-Nucleo ide Poly-
mo phism; SNV: Single-Nucleo ide Va ian ; TF: T ansc ip ion Fac o ; TSS: T ansc ip ion S a Si e; UTR: Un ansla ed Region; VCP:
Va ian -Calling Pipeline; QC: Quali y Con ol
Addi ional Suppo ing In o ma ion may be ound in he online e sion o his a icle.
This is an open access a icle unde he e ms o he C ea i e Commons A ibu ion-NonComme cial-NoDe i a i es License, which pe mi s use and
dis ibu ion in any medium, p o ided he o iginal wo k is p ope ly ci ed, he use is non-comme cial and no modifica ions o adap a ions a e made.
The geno yping o a ian s wi h Sequenom was pe o med by he Technology Cen e, Ins i u e o Molecula Medicine (FIMM), Uni e si y o
Helsinki, Finland
The PRACTICAL Conso ium: Rosalind Eeles: The Ins i u e o Cance Resea ch, 15 Co swold Road, Su on, Su ey SM2 5NG, Uni ed Kingdom.
Royal Ma sden NHS Founda ion T us , Fulham and Su on, London and Su ey, Uni ed Kingdom. Doug Eas on: Depa men o Public Heal h and
P ima y Ca e, Cen e o Cance Gene ic Epidemiology, Uni e si y o Camb idge, S angeways Labo a o y, Wo s Causeway, Camb idge, Uni ed
Kingdom. Kenne h Mui : Uni e si y o Wa wick, Co en y, Uni ed Kingdom. G aham Giles: Cance Epidemiology Cen e, The Cance Council Vic-
o ia, 1 Ra hdowne s ee , Ca l on Vic o ia, Aus alia. Cen e o Molecula , En i onmen al, Gene ic and Analy ic Epidemiology, The Uni e si y o
Melbou ne, Vic o ia, Aus alia. F ed ik Wiklund and Hen ik G €
onbe g: Depa men o Medical Epidemiology and Bios a is ics, Ka olinska Ins i u e,
S ockholm, Sweden. Ch is ophe Haiman: Depa men o P e en i e Medicine, Keck School o Medicine, Uni e si y o Sou he n Cali o nia/No is
Comp ehensi e Cance Cen e , Los Angeles, CA. Johanna Schleu ke : Depa men o Medical Biochemis y and Gene ics, Uni e si y o Tu ku,
Tu ku, Finland. BioMediTech, Uni e si y o Tampe e and FimLab Labo a o ies, Tampe e, Finland. Ma en Weische : Depa men o Clinical Bio-
chemis y, He le Hospi al, Copenhagen Uni e si y Hospi al, He le Ring ej 75, DK-2730 He le , Denma k. Ru h C. T a is: Nu field Depa men o
Clinical Medicine, Cance Epidemiology Uni , Uni e si y o Ox o d, Ox o d, Uni ed Kingdom. Da id Neal: Su gical Oncology (U o-Oncology: S4),
Uni e si y o Camb idge, Box 279, Addenb ooke’s Hospi al, Hills Road, Camb idge, Uni ed Kingdom and Cance Resea ch UK Camb idge Resea ch
Ins i u e, Li Ka Shing Cen e, Camb idge, Uni ed Kingdom. Paul Pha oah: Depa men o Oncology, Cen e o Cance Gene ic Epidemiology, Uni-
e si y o Camb idge, S angeways Labo a o y, Wo s Causeway, Camb idge, Uni ed Kingdom. Kay-Tee Khaw: Camb idge Ins i u e o Public
Heal h, Uni e si y o Camb idge, Fo ie Si e, Robinson Way, Camb idge CB2 0SR, Uni ed Kingdom. Jane L. S an o d: Di ision o Public Heal h Sci-
ences, F ed Hu chinson Cance Resea ch Cen e , Sea le, WA. Depa men o Epidemiology, School o Public Heal h, Uni e si y o Washing on,
Sea le, WA. William J. Blo : In e na ional Epidemiology Ins i u e, 1455 Resea ch Bl d., Sui e 550, Rock ille, MD. S ephen Thibodeau: Mayo Clinic,
Roches e , MN. Ch is iane Maie : Depa men o U ology, Uni e si y Hospi al Ulm, Ge many. Ins i u e o Human Gene ics Uni e si y Hospi al
Ulm, Ge many. Adam S. Kibel: B igham and Women’s Hospi al/Dana-Fa be Cance Ins i u e, 45 F ancis S ee , ASB II-3, Bos on, MA. Washing on
Uni e si y, S Louis, MO. Ceza y Cybulski: Depa men o Gene ics and Pa hology, In e na ional He edi a y Cance Cen e , Pome anian Medical
Uni e si y, Szczecin, Poland. Lisa Cannon-Alb igh : Di ision o Gene ic Epidemiology, Depa men o Medicine, Uni e si y o U ah School o Medi-
cine, Sal Lake Ci y, UT. He mann B enne : Di ision o Clinical Epidemiology and Aging Resea ch, Ge man Cance Resea ch Cen e , Heidelbe g
Ge many. Jong Pa k: Di ision o Cance P e en ion and Con ol, H. Lee Mo fi Cance Cen e , 12902 Magnolia D ., Tampa, FL. Radka Kane a:
Molecula Medicine Cen e and Depa men o Medical Chemis y and Biochemis y, Medical Uni e si y—Sofia, 2 Zd a e S , 1431 Sofia, Bulga ia.
Jyo nsa Ba a: Aus alian P os a e Cance Resea ch Cen e-Qld, Ins i u e o Heal h and Biomedical Inno a ion and Schools o Li e Science and Public
Heal h, Queensland Uni e si y o Technology, B isbane, Aus alia. Manuel R. Teixei a: Depa men o Gene ics, Po uguese Oncology Ins i u e,
Po o, Po ugal and Biomedical Sciences Ins i u e (ICBAS), Po o Uni e si y, Po o, Po ugal. Zsofia Ko e-Ja ai: The Ins i u e o Cance Resea ch,15
Co swold Road, Su on, Su ey SM2 5NG, Uni ed Kingdom. Ali Amin Al Olama and Sa a Benlloch: Uni e si y o Camb idge, S angeways Labo a-
o y, Wo s Causeway, Camb idge, Uni ed Kingdom.
Cance Gene ics
In . J. Cance : 136, 2316–2327 (2015) V
C2014 The Au ho s. Published by Wiley Pe iodicals, Inc. on behal o UICC
In e na ional Jou nal o Cance
IJC
The 2q37 and 17q12-q22 loci a e linked o an inc eased p os a e cance (P Ca) isk. No candida e gene has been localized a
2q37 and he HOXB13 a ian G84E only pa ially explains he linkage o 17q21-q22 obse ed in Finland. We sc eened hese
egions by a ge ed DNA sequencing o sea ch o cance -associa ed a ian s. Al oge he , ou no el suscep ibili y alleles we e
iden i ied. Two ZNF652 (17q21.3) a ian s, s116890317 and s79670217, inc eased he isk o bo h spo adic and he edi a y
P Ca ( s116890317: OR 53.3–7.8, p50.003–3.3 310
25
; s79670217: OR 51.6–1.9, p50.002–0.009). The HDAC4 (2q37.2)
a ian s73000144 (OR 514.6, p50.018) and he EFCAB13 (17q21.3) a ian s118004742 (OR 51.8, p50.048) we e o e -
ep esen ed in pa ien s wi h amilial P Ca. To map he a ian s wi hin 2q37 and 17q11.2-q22 ha may egula e P Ca-
associa ed genes, we combined DNA sequencing esul s wi h ansc ip ome da a ob ained by RNA sequencing. This exp ession
quan i a i e ai locus (eQTL) analysis iden i ied 272 single-nucleo ide polymo phisms (SNPs) possibly egula ing six genes
ha we e di e en ially exp essed be ween cases and con ols. In a modi ied app oach, p e il e ed P Ca-associa ed SNPs we e
exploi ed and in e es ingly, a no el eQTL a ge ing ZNF652 was iden i ied. The no el a ian s iden i ied in his s udy could be
u ilized o P Ca isk assessmen , and hey u he alida e he sugges ed ole o ZNF652 as a P Ca candida e gene. The egu-
la o y egions disco e ed by eQTL mapping inc ease ou unde s anding o he ela ionship be ween egula ion o gene exp es-
sion and suscep ibili y o P Ca and p o ide a aluable s a ing poin o u u e unc ional esea ch.
A la ge p opo ion o amilial p os a e cance (P Ca) cases
can be explained by gene ic isk ac o s.
1
Despi e ex ensi e
esea ch, he iden ifica ion o hese ac o s has p o en chal-
lenging. In Finland, mu a ions in he edi a y p os a e cance
(HPC) isk genes a e ela i ely a e, wi h he excep ion o he
HOXB13 G84E mu a ion,
2
which is p esen in 8.4% o ami-
lial P Ca cases and has been significan ly associa ed wi h an
inc eased P Ca isk in unselec ed cases.
3
The in ol emen o ch omosomal egions 2q37 and
17q12-q22 wi h P Ca has been p e iously epo ed in nume -
ous linkage
4–6
and genome-wide associa ion s udies
(GWASs).
7,8
C opp e al.
9
pe o med a genome-wide linkage
scan o 69 Finnish high- isk HPC amilies and in he domi-
nan model, he loci on 2q37.3 and 17q21-q22 exhibi ed he
s onges linkage signals. No known P Ca candida e gene
esides on 2q37.3, and as demons a ed in ou ea lie s udy,
he HOXB13 G84E mu a ion only pa ially explains he
obse ed linkage o 17q21-q22.
3
He e, we pe o med a ge ed esequencing ha co e ed
he linkage peaks on 2q37 and 17q11.2-q22. The sequence
da a we e fil e ed o iden i y he a ian s wi hin genes p e-
dic ed o be in ol ed in P Ca p edisposi ion. These a ian s
we e alida ed in Finnish HPC amilies and in unselec ed
P Ca pa ien s by Sequenom geno yping, and se e al no el
a ian s we e disco e ed ha we e significan ly associa ed
wi h P Ca. To s udy he impac o single-nucleo ide polymo -
phisms (SNPs) on he egula ion o gene exp ession wi hin
he wo linked egions, we pe o med ansc ip ome sequenc-
ing ollowed by exp ession quan i a i e ai loci (eQTL)
mapping. eQTLs a e known o modi y he pene ance o a e
Wha ’s new?
P os a e cance uns in amilies, bu i s he i abili y isn’ comple ely explained by he gene ic a ian s iden i ied o da e. In
his pape , he au ho s del e deepe in o wo loci ha ha e been linked o p os a e cance . Sequencing da a e ealed ou
new alleles wi hin hese loci ha co ela e wi h inc eased p os a e cance isk. The au ho s hen used he eQTL mapping ech-
nique o iden i y six genes which may be egula ed by a ian s wi hin hese wo loci, genes which had no p e iously been
associa ed wi h p os a e cance .
G an sponso : Academy o Finland; G an numbe : 251074; G an sponso : The Finnish Cance O ganisa ions, he Sig id Juselius
Founda ion and he Compe i i e S a e Resea ch Financing o he Expe Responsibili y A ea o Tampe e Uni e si y Hospi al; G an numbe :
X51003; G an sponso : Eu opean Commission’s Se en h F amewo k P og amme (The PRACTICAL conso ium); G an numbe :
HEALTH-F2–2009-223175; G an sponso : Cance Resea ch UK; G an numbe s: C5047/A7357, C1287/A10118, C5047/A3354, C5047/
A10692 and C16913/A6135; G an sponso : The Na ional Ins i u es o Heal h (Cance Pos -Cance GWAS ini ia i e); G an numbe : 1
U19 CA 148537-01
DOI: 10.1002/ijc.29276
His o y: Recei ed 17 Ap 2014; Accep ed 1 Oc 2014; Online 21 Oc 2014
Co espondence o: Johanna Schleu ke , Medical Biochemis y and Gene ics, Ins i u e o Biomedicine, Kiinamyllynka u 10, Uni e si y o
Tu ku, FI-20014 Tu ku, Finland. Tel.: 1358-2-3337453, Fax: 1358-2-2301280, E-mail: Johanna.Schleu ke @u u.fi
Cance Gene ics
Lai inen e al.2317
In . J. Cance : 136, 2316–2327 (2015) V
C2014 The Au ho s. Published by Wiley Pe iodicals, Inc. on behal o UICC
dele e ious a ian s and he e o e likely con ibu e o gene ic
p edisposi ion o complex diseases. New in o ma ion was
ob ained on se e al genes as well as hei egula o y elemen s
ha gene a ed esh insigh s in o P Ca suscep ibili y, espe-
cially in HPC.
Ma e ial and Me hods
All o he subjec s we e o Finnish o igin. The samples we e
collec ed wi h w i en and signed in o med consen . The can-
ce diagnoses we e confi med using medical eco ds and he
annual upda e om he Finnish Cance Regis y. The p ojec
was app o ed by he local esea ch e hics commi ee a Pi -
kanmaa Hospi al Dis ic and by he Na ional Supe iso y
Au ho i y o Wel a e and Heal h.
Ta ge ed esequencing o 2q37 and 17q11.2-q22
Based on he linkage analysis esul s om C opp e al.,
9
63
P Ca pa ien s and fi e una ec ed indi iduals belonging o 21
Finnish high- isk HPC amilies
10
we e selec ed o a ge ed
esequencing o he 2q37 and 17q11.2-q22 egions (Suppo -
ing In o ma ion Table S1). Each amily had a leas h ee
fi s - o second-deg ee ela i es diagnosed wi h P Ca. Pai ed-
end nex gene a ion sequencing was pe o med a he Tech-
nology Cen e, Ins i u e o Molecula Medicine Finland
(FIMM), Uni e si y o Helsinki. The sequenced agmen s
spanned app oxima ely 6.8 Mb o ch omosome 2q and 21.6
Mb o 17q. The a ge egions we e cap u ed using SeqCap
EZ Choice a ay p obes (Roche NimbleGen, Madison, WI)
and we e sequenced on a Genome Analyze IIx (Illumina,
San Diego, CA) ollowing he manu ac u e ’s p o ocol. The
ead alignmen and a ian calling we e pe o med acco ding
o FIMM’s Va ian -Calling Pipeline (VCP).
11
Bioin o ma ics wo k low o a ian cha ac e iza ion
A schema ic o e iew o ou bioin o ma ics wo kflow is
shown in Figu e 1. Only hose a ian s ha we e p esen in
all he a ec ed amily membe s we e selec ed o subsequen
analysis. The a ian s we e anno a ed using Ensembl V65
gene se e ie ed om he UCSC Genome B owse .
12
The
pheno ypic e ec s o he a ian s we e s udied wi h h ee in
silico pa hogenici y p edic ion p og ams. Mu a ionTas e
13
classifies single-nucleo ide a ian s (SNVs) and small inse -
ion/dele ion polymo phisms (indels) as polymo phic o
pa hogenic. PolyPhen-2
14
and PON-P
15
only p edic he
e ec s o nonsynonymous SNVs ha esul in amino acid
eplacemen . PolyPhen-2 classifies he a ian s as benign,
possibly pa hogenic o p obably pa hogenic, whe eas PON-P
defines hem as neu al, unclassified o pa hogenic. Va ian s
ca ego ized as pa hogenic by a leas one ole ance p edic o
we e defined as pa hogenic. In addi ion, mino allele equen-
cies (MAFs) we e ob ained om he dbSNP da abase and
in o ma ion on known P Ca-associa ed genes was e ie ed
om he COSMIC
16
and DDPC
17
da abases. Pa hway da a
we e ga he ed om Pa hway Commons,
18
KEGG
19
and
WikiPa hways
20
and Gene On ology da a we e e ie ed om
Ensembl BioMa .65.
21
Highe p io i y was assigned o a e
a ian s (MAF <0.05), a ian s loca ed in genes p e iously
linked o P Ca, and a ian s loca ed in genes unc ionally
simila o P Ca-associa ed genes.
Valida ion o p edic ed P Ca-associa ed a ian s wi h
Sequenom
A e fil e ing, 58 a ian s in 35 a ge genes (lis ed in Sup-
po ing In o ma ion Tables S2–S4) we e selec ed o alida-
ion which was pe o med on ge mline DNA om 2,216
subjec s, including 1,293 cases and 923 popula ion con ols.
The majo i y o he cases (1,105 indi iduals) ep esen ed
Figu e 1. A lowcha desc ibing he a ian cha ac e iza ion pipe-
line. The a ge ed esequencing o 2q37 and 17q11.2-q22 om 68
Finnish HPC amily membe s p oduced a o al o 107,479 unique
sequence a ian s. Family-based il e ing excluded 66,867 a ian s
ha did no coseg ega e wi h a ec ion s a us. Anno a ion enabled
he selec ion o 24,813 a ian s ha we e loca ed wi hin p o ein-
coding genes. Pa hogenici y p edic ions we e pe o med in silico
using Mu a ionTas e , PolyPhen-2 and PON-P. As a esul , he num-
be o candida e a ian s was educed o 152. The inal il e ing
s ep exploi ed di e se in o ma ion on genes and a ian s as well
as gene on ology and pa hway da a s o ed in se e al public da a-
bases. In addi ion, selec HDAC4, ZNF652 and HOXB13 a ian s,
which we e p edic ed o be nonpa hogenic, we e included in he
alida ion because hese genes ha e been associa ed wi h P Ca in
p e ious s udies.
Cance Gene ics
2318 Fine-mapping he 2q and 17q p os a e cance loci
In . J. Cance : 136, 2316–2327 (2015) V
C2014 The Au ho s. Published by Wiley Pe iodicals, Inc. on behal o UICC
unselec ed P Ca pa ien s om he Pi kanmaa Hospi al Dis-
ic , Tampe e, Finland. In addi ion, 188 index cases om
Finnish HPC amilies
10
we e included in he s udy. The con-
ol DNA samples om anonymous male blood dono s we e
p o ided by he Finnish Red C oss Blood T ans usion Se -
ice. Geno yping was pe o med a he Technology Cen e,
FIMM using he Sequenom MassARRAY sys em and iPLEX
Gold assays (Sequenom, San Diego, CA). Geno yping eac-
ions we e pe o med wi h 20 ng o d ied genomic DNA
acco ding o manu ac u e ’s ecommenda ions and wi h hei
eagen s. The geno ypes we e called using Type Analyze
so wa e (Sequenom). Fo quali y con ol (QC) easons, he
geno ype calls we e also checked manually. Geno yping qual-
i y was examined using a de ailed QC p ocedu e ha
included success a e checks, duplica ed samples and wa e
con ols.
S a is ical and bioin o ma ic analyses o he alida ed
a ian s
Associa ion and Ha dy-Weinbe g equilib ium (HWE) es s
we e pe o med using PLINK.
22
The p alue h eshold o
he HWE es was se o 0.05. Samples wi h low geno yping
equencies (<0.80) we e excluded om he associa ion anal-
ysis. The s a is ical significance o he associa ion was e al-
ua ed using a wo-sided Fishe ’s exac es . Odds a ios (OR)
we e calcula ed using PLINK wi h op ion — fishe . No u -
he model adjus men s o con ounding ac o s we e made.
ENCODE in o ma ion
23
o noncoding a ian s was e ie ed
om he Regulome da abase (RegulomeDB).
24
The linkage
disequilib ium (LD) analysis o he s a is ically significan
a ian s is desc ibed in Supplemen a y Me hods.
Geno yping o he op ou candida e a ian s in Finnish
HPC amilies
Fou a ian s we e chosen o seg ega ion analysis in Finnish
HPC amilies based on a s ong associa ion wi h P Ca, a
high OR alue and/o p edic ed pa hogenici y. The coseg e-
ga ion o s116890317 and s79670217 in ZNF652 (Re Seq
NM_001145365), s73000144 in HDAC4 (Re Seq
NM_006037) and s118004742 in EFCAB13 (Re Seq
NM_152347) wi h a ec ion s a us was de e mined in 41 am-
ilies whose index cases we e mu a ion-posi i e in he Seque-
nom alida ion. Fo hese amilies, DNA samples we e
a ailable om 243 P Ca cases and 204 heal hy amily mem-
be s. The a ian s we e geno yped in wo o 17 (median:
se en) indi iduals pe amily by Sange sequencing.
RNA ex ac ion and sequencing
Pe iphe al blood samples collec ed in PAXgene
V
R
Blood RNA
Tubes (P eAnaly iX GmbH, Swi ze land) we e a ailable om
84 P Ca pa ien s and 15 heal hy male ela i es belonging o
31 Finnish HPC amilies. These included 11 amilies om
he a ge ed esequencing s ep (Suppo ing In o ma ion Table
S1) and addi ional 20 high- isk amilies.
10
To al RNA was
pu ified wi h MagMAX
TM
o S abilized Blood Tubes RNA
Isola ion Ki (Ambion
V
R
/Li e Technologies, Ca lsbad, CA) and
wi h a PAXgene Blood miRNA Ki (P eAnaly iX GmbH).
RNA in eg i y and quali y we e analyzed using he Agilen
2100 Bioanalyze and he Agilen RNA 6000 Nano Ki (Agi-
len Technologies, San a Cla a, CA). The massi ely pa allel
pai ed-end RNA sequencing was pe o med a Beijing
Genomics Ins i u e (BGI Hong Kong Co., Tai Po, Hong
Kong) using an Illumina HiSeq2000 sequencing pla o m
(Illumina).
RNA sequencing da a analysis
On a e age, RNA sequencing p oduced 45 million eads pe
sample. The QC check was pe o med using as QC (h p://
www.bioin o ma ics.bbs c.ac.uk/p ojec s/ as qc). The eads
we e aligned wi h Topha 2
25
using GRCh37/hg19 as he e -
e ence genome. The ead coun s o he genes we e de e -
mined using HTSeq (h p://www-hube .embl.de/use s/ande s/
HTSeq/). The aw ead coun s we e ans o med in o compa-
able exp ession alues ia no maliza ion using he DESeq
package o R
26
and he genes wi h e y low o no exp ession
(no malized ead coun s o <20) we e emo ed. A di e en-
ial gene exp ession analysis was hen pe o med using a
wo-sided Mann–Whi ney es wi h a p alue cu o o 0.05.
eQTL mapping and da a analysis
The eQTL analysis was based on he RNA-seq da a and on
he SNP geno ypes ob ained om a ge ed DNA sequencing.
This da a exis ed o 19 samples a 2q37 and o 17 samples
a 17q11.2-q22. In o al, 54,919 SNPs (a e age 6,865 pe
gene, see Suppo ing In o ma ion Table S5 o de ails) we e
es ed o associa ion wi h hei candida e a ge genes. Only
genes wi h di e en ial exp ession (DE) pa e ns be ween
heal h s a us g oups we e included in he eQTL analysis, o
inc ease he p obabili y ha ound SNP-gene associa ions
also link P Ca wi h a ce ain SNP geno ype. The eQTL map-
ping was applied on 2q37 and 17q11.2-q22 o iden i y cis-
egula ed genes. SNPs associa ed in cis we e defined as a -
ian s loca ed wi hin 1 Mb up- o downs eam o he gene
unde s udy. The significance le el o SNP-gene associa ions
was se o p0.005. A mul iple es ing adjus men was omi -
ed because o he la ge numbe o es ed SNPs and he
na u e o he pe mu a ion ype es s, acknowledging ha his
esul ed in comp omised esolu ion.
A modified cis-eQTL app oach was also u ilized, whe ein
a la ge geno ype da ase om he iCOGS s udy
27
was used o
p eiden i y possible P Ca-associa ed SNPs o 2,824 unse-
lec ed Finnish P Ca pa ien s and 2,440 con ols. He e, Fish-
e ’s exac es wi h a modes significance le el o 0.005 was
used o s udy he associa ion. Significan iCOGS a ian s ha
we e also obse ed in he a ge ed DNA sequencing da a
we e hen selec ed o eQTL analysis, which was es ic ed o
he fine-mapped egions. Addi ional de ails o he eQTL
analysis a e p esen ed in Supplemen a y Me hods.
RegulomeDB was used o anno a e and assess he egula-
o y po en ial o he de ec ed eQTLs.
24
The ENCODE
Cance Gene ics
Lai inen e al.2319
In . J. Cance : 136, 2316–2327 (2015) V
C2014 The Au ho s. Published by Wiley Pe iodicals, Inc. on behal o UICC
da ase s
23
we e e ie ed om he UCSC Genome B owse
websi e o isualiza ion pu poses using he Table B owse
ool.
12
As a gene al indica o o egula o y po en ial, we used
he da ase ha con ained en iched DNase hype sensi i e
si es in 125 cell ypes. To highligh he egula o y po en ial o
eQTLs in P Ca issue, we used he LNCaP DNase (wgEnco-
deAwgDnaseUwDukeLncapUniPk) and LNCaP (And )
DNase (wgEncodeAwgDnaseUwDukeLncapand ogenUniPk)
da ase s con aining DNase hype sensi i e si es in LNCaP cells
unde no mal and and ogen-induced condi ions, espec i ely.
T ansc ip ion ac o (TF) binding si e da a we e ga he ed
om he Txn Fac ChIP V3 da ase , which con ains ChIP-seq
expe imen al da a on 91 cell ypes and 189 TFs.
Resul s
Ta ge ed DNA sequencing da a analysis
The pe cen age o mapped eads was 95.0 and 95.7% o he
samples sequenced o 2q37 and 17q11.2-q22, espec i ely. The
a ge co e age was 99.8% o 2q37 and 99.5% o 17q11.2-q22.
Co espondingly, he pe cen age o bases ha ing co e age o
203o mo e was 79.9 and 63.4%. The o al numbe o unique
a ian s ac oss all samples disco e ed by he u ilized VCP was
107,479 (Fig. 1). Among he 41 p edic ed pa hogenic a ian s
in 2q37, he e we e 20 missense SNVs, 16 noncoding SNVs
and fi e indels. O all 111 p edic ed pa hogenic a ian s in
17q11.2-q22, wo a ian s we e nonsense SNVs, 49 we e mis-
sense SNVs, 36 we e noncoding SNVs and 24 we e indels.
P Ca-associa ed a ian s iden i ied by Sequenom
alida ion
Following p io i iza ion, a o al o 58 a ian s we e selec ed
o alida ion in a la ge sample se (Suppo ing In o ma ion
Table S2). In he QC analysis, ou a ian s ailed he HWE
es (p<0.05), and 20 samples we e omi ed due o low gen-
o yping equencies (<0.80). In he case-con ol associa ion
analysis, a o al o 13 a ian s in se en di e en genes we e
s a is ically significan ly associa ed wi h P Ca (p<0.05;
Tables 1 and 2 and Suppo ing In o ma ion Tables S3 and
S4). Th ee a ian s we e loca ed in he ZNF652 gene a
17q21.3, and he HDAC4 (2q37.2), HOXB3 (17q21.3),
ACACA (17q21) and MYEOV2 (2q37.3) genes ha bo ed wo
a ian s each. A single a ian was iden ified in he HOXB13
and EFCAB13 genes a 17q21.3. Only h ee o hese 13 P Ca-
associa ed a ian s we e loca ed wi hin exons, whe eas he
majo i y, 10 a ian s, esided in noncoding egions.
Fou o he a ian s wi h a s a is ically significan associa-
ion wi h P Ca we e p esen in bo h he amilial and he
unselec ed sample se s. These we e s116890317 and
s79670217 in ZNF652, s10554930 in HOXB3 and
s13411615 in MYEOV2. The wo ZNF652 a ian s had he
s onges associa ion wi h an inc eased P Ca isk.
s116890317 had he mos significan associa ion wi h he
amilial cases (OR 57.8, 95% CI 3.0–20.3, p53.3 310
25
)
and he same a ian con e ed he highes isk o 3.3 (95%
CI 1.4–7.5, p50.003) among he unselec ed cases.
s79670217 had he mos significan associa ion wi h P Ca in
he unselec ed sample se (p50.002) and was he second
mos significan a ian in he amilial P Ca pa ien s
(OR 51.9, 95% CI 1.2–3.1, p50.009; Tables 1 and 2).
The highes OR o 14.6 (95% CI 1.5–140.2, p50.018) was
obse ed o he HDAC4 a ian s73000144 (c.958C>T,
p.Val320Ile) among he amilial samples (Table 1). Only h ee
amilial P Ca pa ien s (1.6%), se en unselec ed pa ien s
(0.6%) and one con ol indi idual (0.1%) ca ied he mino
allele in a he e ozygous s a e, and none o he geno yped
indi iduals we e homozygous. s73000144 was p edic ed o
Table 1. Va ian s signi ican ly associa ed wi h p os a e cance based on a compa ison o amilial cases (n5186) and con ols (n5914)
SNP Id Func ion Gene Ch Min/Maj F_A/F_U (%) p alue OR (95% CI) Pa hogenici y p edic ion
s116890317 In onic ZNF652 17 A/T 2.96/0.39 3.3 310
25
7.8 (3.0–20.3) Polymo phism/–/–
s79670217 In onic ZNF652 17 G/T 6.65/3.56 0.009 1.9 (1.2–3.1) Polymo phism/–/–
s10554930 In onic HOXB3 17 2ACA/ACA 27.5/21.3 0.010 1.4 (1.1–1.8) Pa hogenic/–/–
s35384813 50-UTR HOXB3 17 1T/– 26.7/20.8 0.013 1.4 (1.1–1.8) Pa hogenic/–/–
s73000144 Missense HDAC4 2 T/C 0.80/0.06 0.018 14.6 (1.5–140.2) Polymo phism/benign/neu al
s13411615
1
Nea gene 50MYEOV2 2 C/A 52.1/45.6 0.023 1.3 (1.0–1.6) Polymo phism/–/–
s9899142 In onic HOXB13 17 T/C 11.2/15.6 0.031 0.7 (0.5–1.0) Polymo phism/–/–
s118004742 Nonsense EFCAB13 17 G/T 4.79/2.73 0.048 1.8 (1.0–3.1) Pa hogenic/–/–
s142044482 30-UTR ZNF652 17 1A/– 2.94/1.59 0.087 1.9 (0.9–3.8) Polymo phism/–/–
s140611363
1
Nea gene 50ACACA 17 2A/A 28.8/31.1 0.421 0.9 (0.7–1.1) Pa hogenic/–/–
s72828246
1
Nea gene 50ACACA 17 G/A 28.8/30.9 0.459 0.9 (0.7–1.2) Pa hogenic/benign/neu al
s13406410
1
Nea gene 50MYEOV2 2 C/T 47.6/46.8 0.817 1.0 (0.8–1.3) Pa hogenic/–/–
s61752234 Synonymous HDAC4 2 C/T 7.22/6.83 0.823 1.1 (0.7–1.6) Polymo phism/–/–
Bold signi ies p<0.05.
1
Va ian s a e in linkage disequilib ium.
Abb e ia ions: Ch : ch omosome; Min: mino allele; Maj: majo allele; F_A: equency o he mino allele in cases; F_U: equency o he mino allele
in con ols; OR: odds a io; CI: con idence in e al; pa hogenici y p edic ion esul s om: Mu a ionTas e /PolyPhen-2/Pon-P.
Cance Gene ics
2320 Fine-mapping he 2q and 17q p os a e cance loci
In . J. Cance : 136, 2316–2327 (2015) V
C2014 The Au ho s. Published by Wiley Pe iodicals, Inc. on behal o UICC

be benign o neu al by all h ee in silico pa hogenici y p e-
dic ion algo i hms (Suppo ing In o ma ion Table S2).
The s118004742 nonsense mu a ion (c.1638T>G,
p.Ty 546Te ) in he EFCAB13 gene was p edic ed o be pa h-
ogenic by Mu a ionTas e (Suppo ing In o ma ion Table S2).
Th ee amilial cases (1.6%) we e homozygous o he mino
allele. The e we e 12 he e ozygo es among he amilial index
cases (6.5%) and 66 among he unselec ed cases (6.0%). A
s a is ically significan associa ion be ween s118004742 and
P Ca was only obse ed o he amilial pa ien s (Table 1).
The OR o 1.8 (95% CI 1.0–3.1) sugges ed an inc eased isk
o HPC. s118004742 ca ie s in he unselec ed sample se
did no ha e an inc eased cance isk (OR 51.1, 95% CI 0.8–
1.6, p50.637; Suppo ing In o ma ion Table S4).
Two common noncoding a ian s in he HOXB3 gene,
s10554930 and s35384813, had a mode a e e ec on P Ca
isk, wi h ORs anging om 1.2 o 1.4 (Tables 1 and 2).
Mu a ionTas e p edic ed bo h o hese a ian s o be pa ho-
genic (Suppo ing In o ma ion Table S2). Fo fi e a ian s,
he ORs we e <1.0, indica ing a modula o y ole in P Ca
p edisposi ion. These a ian s we e loca ed nea o wi hin
he ZNF652, HDAC4, HOXB13 and ACACA genes (Tables 1
and 2). Acco ding o he RegulomeDB, h ee o he 13 s a is-
ically significan a ian s we e likely o a ec p o ein bind-
ing: s9899142 in HOXB13 (Regulome sco e o 1 ),
s13406410 in MYEOV2 and s72828246 in ACACA (bo h
ha ing Regulome sco e o 2b).
In case-case compa isons, none o he iden ified a ian s
we e significan ly associa ed wi h Gleason sco e, a e age age
o he se um p os a e specific an igen (PSA) le el a diagnosis
(da a no shown). The LD analysis (Suppo ing In o ma ion
Fig. S1) e ealed ha none o ou 13 s a is ically significan
a ian s (Tables 1 and 2) we e in linkage disequilib ium wi h
p e iously epo ed P Ca-associa ed a ian s
27
(see Supple-
men a y Resul s o de ails).
Seg ega ion analysis o he op ou candida e a ian s
Al oge he , 41 amilial index cases ou o 188 geno yped by
Sequenom ca ied a leas one o he op ou candida e a -
ian s. Seg ega ion analysis was pe o med o hese 41 HPC
amilies. s116890317, s79670217 and s118004742 we e
mo e common among P Ca pa ien s han heal hy amily
membe s and p o ided e idence o coseg ega ion wi h a ec-
ion s a us in 20 amilies (Suppo ing In o ma ion Tables S6–
S8). Howe e , in 15 o hese amilies, una ec ed male mu a-
ion ca ie s we e also obse ed. In se en amilies, all o he
una ec ed male ca ie s we e young enough (<55 yea s) o
de elop P Ca la e in li e. s116890317 seg ega ed comple ely
wi h a ec ion s a us in one amily (Suppo ing In o ma ion
Fig. S2a), as did s79670217 (Suppo ing In o ma ion Fig.
S2b). Comple e seg ega ion o s118004742 was obse ed in
h ee amilies (Suppo ing In o ma ion Table S8). The
HDAC4 a ian s73000144 was de ec ed in h ee amilies,
and app oxima ely one- hi d o he amily membe s we e
iden ified as ca ie s, i espec i e o hei heal h s a us (Sup-
po ing In o ma ion Table S9).
Mul iple a ian s we e obse ed in 16 indi iduals om 14
amilies. Two amilies ha bo ed s116890317, s79670217 and
s118004742, whe eas one amily was posi i e o
s79670217, s73000144 and s118004742. In he emaining
amilies, he mos common combina ion de ec ed was
s79670217 oge he wi h s118004742 (six amilies). E i-
dence o seg ega ion wi h a ec ion s a us was ob ained o a
maximum o one a ian pe amily.
Table 2. Va ian s signi ican ly associa ed wi h p os a e cance based on a compa ison o unselec ed cases (n51096) and con ols (n5914)
SNP Id Func ion Gene Ch Min/Maj F_A/F_U (%) p alue OR (95% CI) Pa hogenici y p edic ion
s79670217 In onic ZNF652 17 G/T 5.66/3.56 0.002 1.6 (1.2–2.2) Polymo phism/–/–
s116890317 In onic ZNF652 17 A/T 1.27/0.39 0.003 3.3 (1.4–7.5) Polymo phism/–/–
s13406410
1
Nea gene 50MYEOV2 2 C/T 51.5/46.8 0.006 1.2 (1.1–1.4) Pa hogenic/–/–
s61752234 Synonymous HDAC4 2 C/T 4.85/6.83 0.008 0.7 (0.5–0.9) Polymo phism/–/–
s142044482 30-UTR ZNF652 17 1A/- 0.68/1.59 0.009 0.4 (0.2–0.8) Polymo phism/–/–
s140611363
1
Nea gene 50ACACA 17 2A/A 27.9/31.1 0.032 0.9 (0.7–1.0) Pa hogenic/–/–
s10554930 In onic HOXB3 17 2ACA/ACA 24.1/21.3 0.034 1.2 (1.0–1.4) Pa hogenic/–/–
s13411615
1
Nea gene 50MYEOV2 2 C/A 49.0/45.6 0.037 1.1 (1.0–1.3) Polymo phism/–/–
s72828246
1
Nea gene 50ACACA 17 G/A 28.0/30.9 0.044 0.9 (0.8–1.0) Pa hogenic/benign/neu al
s35384813 50-UTR HOXB3 17 1T/– 23.2/20.8 0.073 1.1 (1.0–1.3) Pa hogenic/–/–
s73000144 Missense HDAC4 2 T/C 0.33/0.06 0.078 5.9 (0.7–47.9) Polymo phism/benign/neu al
s118004742 Nonsense EFCAB13 17 G/T 3.0/2.7 0.637 1.1 (0.8–1.6) Pa hogenic/–/–
s9899142 In onic HOXB13 17 T/C 16.1/15.6 0.665 1.0 (0.9–1.2) Polymo phism/–/–
Bold signi ies p<0.05.
1
Va ian s a e in linkage disequilib ium.
Abb e ia ions: Ch : ch omosome; Min: mino allele; Maj: majo allele; F_A: equency o he mino allele in cases; F_U: equency o he mino allele
in con ols; OR: odds a io; CI: con idence in e al; pa hogenici y p edic ion esul s om: Mu a ionTas e /PolyPhen-2/Pon-P.
Cance Gene ics
Lai inen e al.2321
In . J. Cance : 136, 2316–2327 (2015) V
C2014 The Au ho s. Published by Wiley Pe iodicals, Inc. on behal o UICC
eQTL mapping esul s
Di e en ial gene exp ession analysis e ealed h ee genes (o
173 es ed) loca ed a 2q37 and fi e genes (o 761 es ed) a
17q11.2-q22 whose exp ession le els di e ed significan ly
be ween cases and con ols (p<0.05). In he a ge ed cis-
eQTL analysis, SNPs wi hin 2 Mb windows we e es ed o
associa ion wi h each o hese eigh DE genes (Suppo ing
In o ma ion Table S5). Al oge he , 272 candida e egula o y
SNPs we e iden ified o six DE genes only (Suppo ing
In o ma ion Table S10). A as majo i y, 237 candida e SNPs
po en ially egula e he exp ession o AGAP1, SCLY and
NDUFA10 a 2q37 (Fig. 2). The emaining 35 candida e
SNPs possibly egula e TBKBP1, PNPO and NAGS a
17q11.2-q22 (Fig. 3). Based on he ENCODE da a, he
s onges e idence o egula o y po en ial was ound o
s11650354 on ch omosome 17, which a ge s he TBKBP1
Figu e 2. Cis-eQTLs a ge ing di e en ially exp essed genes on ch omosome 2. All s a is ically signi ican eQTLs a e indica ed wi h a ack o
black ba s. Selec ed eQTLs, s12620966 and s983221 ( a ge ing AGAP1) and s1996513 and s12712297 ( a ge ing NDUFA10) a e illus-
a ed in mo e de ail. DNaseI hype sensi i e si es om he DNase clus e and LNCaP da ase s a e indica ed wi h g een and ed ec angles,
espec i ely. Blue ec angles deno e TF binding si es.
Cance Gene ics
2322 Fine-mapping he 2q and 17q p os a e cance loci
In . J. Cance : 136, 2316–2327 (2015) V
C2014 The Au ho s. Published by Wiley Pe iodicals, Inc. on behal o UICC
gene. This known eQTL o e laps wi h an open ch oma in
egion (Mc 7 and Gm12892 cell lines) and i s ole in he eg-
ula ion o TBKBP1 exp ession has been confi med in a p e i-
ous s udy.
28
s12620966 a ge ing AGAP1 on ch omosome 2
o e laps wi h se e al TF binding si es disco e ed by ChIP-
seq (HepG2 cell line), posi ion weigh ma ix (PWM) ma ch-
ing and digi al DNaseI oo p in ing s udies (Suppo ing
In o ma ion Table S10). None o he coding a ian s ha
we e iden ified by a ge ed DNA sequencing and alida ed by
Sequenom we e s a is ically significan eQTLs (da a no
shown).
The modified cis-eQTL analysis was based on 12 SNPs a
2q37 and 22 SNPs a 17q11.2-q22 ha we e sha ed be ween
he iCOGS da ase and ou se o a ian s ob ained by a -
ge ed esequencing. The egula o y po en ial o hese 34
SNPs was e alua ed o 144 genes a 2q37 and o 160 genes
Figu e 3. Cis-eQTLs a ge ing di e en ially exp essed genes on ch omosome 17. All s a is ically signi ican eQTLs a e indica ed wi h a ack
o black ba s. Selec ed eQTLs, s11650354 ( a ge ing TBKBP1) and s12951323 ( a ge ing PNPO) a e illus a ed in mo e de ail. DNaseI
hype sensi i e si es om he DNase clus e and LNCaP da ase s a e indica ed wi h g een and ed ec angles, espec i ely. Blue ec angles
deno e TF binding si es.
Cance Gene ics
Lai inen e al.2323
In . J. Cance : 136, 2316–2327 (2015) V
C2014 The Au ho s. Published by Wiley Pe iodicals, Inc. on behal o UICC
a 17q11.2-q22. The modified eQTL app oach iden ified only
one P Ca-associa ed candida e eQTL on ch omosome 2 and
36 candida e eQTLs on ch omosome 17. Selec ed examples
o hese eQTLs and hei a ge genes a e shown in Suppo -
ing In o ma ion Table S11. The ENCODE da a om Regulo-
meDB indica ed he s onges e idence o egula o y
po en ial o wo a ian s on ch omosome 17, s4796751 and
s4796616, which a ge he DHX58, MLX and JUP genes.
Bo h a ian s ha e p e iously been epo ed as eQTLs a ge -
ing MGC20781 and NT5C3L
29
and hey o e lap wi h open
ch oma in egions (in 16 and 17 cell lines, espec i ely).
s4796616 is also loca ed wi hin a TF binding si e (U2OS cell
line). Two addi ional ch omosome 17 a ian s, s4793943
and s16941107 we e defined as likely o a ec gene exp es-
sion. These a ian s a ge he ZNF652 and ARL17B genes,
espec i ely, and o e lap wi h open ch oma in egions (in 6
and 42 cell lines, espec i ely) as well as se e al TF binding
si es (Suppo ing In o ma ion Table S11). O pa icula in e -
es was he ch omosome 17 a ian s4793976 a ge ing he
SPOP gene. Al hough no da a o his eQTL was a ailable in
he RegulomeDB, he impo ance o SPOP in P Ca p edispo-
si ion has been ecognized.
30
Discussion
P io s udies ha e iden ified a s ong ela ionship be ween
P Ca and linkage o ch omosomal egions 2q37 and 17q11.2-
q22. Inspi ed by he lack o candida e genes and mu a ions,
we esequenced he linkage peaks and confi med he
sequencing esul s by alida ing selec a ian s. As he num-
be o a ian s p o ided by he VCP was high, hei p io i i-
za ion o alida ion was c i ical.
The a ian s ha we e s a is ically significan ly associa ed
wi h P Ca we e clus e ed in wo genes on ch omosome 2q37,
HDAC4 and MYEOV2, and in fi e genes on ch omosome
17q11.2-q22, ZNF652, HOXB3, HOXB13, EFCAB13 and
ACACA (Tables 1 and 2). In e es ingly, ou o hese genes,
HDAC4, ZNF652, HOXB3 and HOXB13 encode TFs. T an-
sc ip ional egula ion plays an essen ial ole in main aining
no mal gene con ol, and mu a ions in genes coding o TFs
ha e been iden ified in P Ca. Examples o commonly occu -
ing al e a ions include he usion o TMPRSS2 wi h ERG,
and mu a ions in genes coding o he o khead-box amily
o TFs.
31
The ZNF652 gene a 17q21.3 codes o a DNA-binding
ansc ip ional ep esso p o ein wi h se en zinc finge
mo i s.
32
Highes exp ession le els ha e been de ec ed in no -
mal b eas , p os a e and panc eas, whe eas in p ima y umo s
and cance cell lines, ZNF652 exp ession is gene ally lowe .
32
Howe e , in P Ca, he coexp ession o high le els o ZNF652
and he and ogen ecep o (AR) has been shown o inc ease
he isk o PSA elapse.
33
In addi ion, he ecen ly cha ac e -
ized ZNF652 DNA binding si e was ound in he p omo e s
o se e al genes ha a e in ol ed in P Ca de elopmen and
p og ession.
34
ZNF652 also in e ac s wi h CBFA2T3, a pu a-
i e b eas cance umo supp esso , which has been shown
o enhance he ep esso ac i i y o ZNF652.
32
To da e, only a single P Ca-associa ed isk a ian has
been iden ified in he ZNF652 gene. s7210100 has been
epo ed o p edispose men o A ican descen o P Ca. The
isk allele is p esen a a low equency (<1%) in non-
A ican popula ions.
35
A possible Eu opean-specific isk a i-
an , s11650494, is loca ed in a lincRNA jus downs eam o
he ZNF652 gene and was ecen ly desc ibed by he PRACTI-
CAL Conso ium.
27
The p esen s udy iden ified wo no el
ZNF652 gene a ian s, s116890317 and s79670217, which
we e significan ly associa ed wi h P Ca in bo h amilial and
unselec ed cases. The isk associa ion was pa icula ly appa-
en in pa ien s wi h a posi i e amily his o y o he disease.
Co espondingly, bo h a ian s showed e idence o a leas
pa ial coseg ega ion wi h a ec ion s a us in a subs an ial
po ion o Finnish HPC amilies. Like s7210100, hese wo
no el a ian s a e loca ed in he fi s in on o he gene, sug-
ges ing ha hey may play a ole in egula ing ZNF652 by
a ec ing splicing e en s and/o issue-specific exp ession.
The HDAC4 gene a 2q37.2 encodes a well-cha ac e ized
ansc ip ional ep esso . HDAC4 has been epo ed o accu-
mula e in he nucleus in ho mone- e ac o y P Ca
36
and o
bind o and inhibi he ac i i y o AR by SUMOyla ion.
37
He e, we de e mined ha he exonic HDAC4 a ian
s73000144 (c.958C>T) was significan ly associa ed wi h am-
ilial P Ca (OR 514.6, 95% CI 1.5–140.2, p50.018). The a -
ian also had a high OR (55.8, 95% CI 0.7–47.9) among he
unselec ed cases (Suppo ing In o ma ion Table S4), sugges -
ing an inc eased cance isk, bu his esul was no s a is i-
cally significan (p50.078). The pa hogenici y o s73000144
is unce ain. The esul ing amino acid change, a subs i u ion
o isoleucine o aline (p.Val320Ile) is conse a i e and was
no conside ed pa hogenic by any o he in silico p edic o s
used (Suppo ing In o ma ion Table S2). The s ikingly high
OR o he amilial sample se , oge he wi h he obse a ion
ha his a ian was de ec ed in only h ee ou o 186 index
cases om he Finnish HPC amilies, sugges ed ha
s73000144 may be a p i a e mu a ion. The impo ance o
p i a e mu a ions has been emphasized in many diseases,
some o which a e associa ed wi h specific e hnic g oups.
The p o ein encoded by he EFCAB13 (EF-hand calcium
binding domain 13) gene a 17q21.3 con ains a pa icula
helix-loop-helix domain, he EF-hand, which is equi ed o
calcium ion binding. EF-hands a e o en ound in calcium
senso and calcium signal modula o p o eins. Ca
21
binding
igge s a con o ma ional change in he EF-hand mo i ,
which leads o he ac i a ion o inac i a ion o a ge p o-
eins. Cu en ly, he e is no e idence linking EFCAB13 wi h
P Ca. The nonsense mu a ion s118004742 in he EFCAB13
gene in oduces a p ema u e s op codon, leading o a signifi-
can unca ion o he nascen p o ein. T unca ing mu a ions
a e gene ally conside ed dele e ious and, as expec ed,
s118004742 was p edic ed pa hogenic by Mu a ionTas e
(Suppo ing In o ma ion Table S2). The a ian seg ega ed
Cance Gene ics
2324 Fine-mapping he 2q and 17q p os a e cance loci
In . J. Cance : 136, 2316–2327 (2015) V
C2014 The Au ho s. Published by Wiley Pe iodicals, Inc. on behal o UICC