Genome sequencing and analysis of Mangalica, a fatty local pig of Hungary

RESEARCH ARTICLE Open Access Genome sequencing and analysis o Mangalica, a a y local pig o Hunga y János Molná 2,3 , Tibo Nagy 1 , Vik o S ége 1 , Gábo Tó h 1,4 , Fe enc Ma incs 1* and End e Ba a 1* Abs ac Backg ound: Mangalicas a e a y ype local/ a e pig b eeds wi h an inc easing p esence in he niche po k ma ke in Hunga y and in o he coun ies. To explo e hei gene ic esou ces, we ha e analysed da a om nex -gene a ion sequencing o an indi idual male om each o h ee Mangalica b eeds along wi h a local male Du oc pig. S uc u al a ia ions, such as SNPs, INDELs and CNVs, we e iden i ied and pa icula genes wi h SNP a ia ions we e analysed wi h special emphasis on unc ions ela ed o a me abolism in pigs. Resul s: Mo e han 60 Gb o sequence da a we e gene a ed o each o he sequenced indi iduals, esul ing in 11× o 19× au osomal median co e age. A e s ingen il e ing, a ound six million SNPs, o which app oxima ely 10% a e no el compa ed o he dbSNP138 da abase, we e iden i ied in each animal. Se e al hund ed housands o INDELs and abou 1,000 CNV gains we e also iden i ied. The unc ional anno a ion o genes wi h exonic, non-synonymous SNPs, which a e common in all h ee Mangalicas bu a e absen in ei he he e e ence genome o he sequenced Du oc o his s udy, highligh ed 52 genes in lipid me abolism p ocesses. Fu he analysis e ealed ha 41 o hese genes a e associa ed wi h lipid me abolic o egula o y pa hways, 49 a e in a -me abolism and a ness-pheno ype QTLs and, wi h he excep ion o ACACA,ANKRD23,GM2A,KIT, MOGAT2, MTTP, FASN, SGMS1, SLC27A6 and RETSAT,ha eno p e iously been associa ed wi h a - ela ed pheno ypes. Conclusions: Genome analysis o Mangalica b eeds e ealed ha local/ a e b eeds could be a ich sou ce o sequence a ia ions no p esen in cosmopoli an/indus ial b eeds. The iden i ied Mangalica a ia ions may, he e o e, be a e y use ul esou ce o u u e s udies o ag onomically impo an ai s in pigs. Keywo ds: Mangalica, Genome sequencing, Fa y pig, B eed-speci ic SNP, Gene unc ion Backg ound Due o he economic alue o a m animals, hei gen- omics, in gene al, and whole genome sequencing, in pa - icula , a e impo an issues. Resul s o such esea ch ha e al eady had an impac and will con inue o do so in he u u e in e ms o p oduc ion o mea , milk, ib e and o he p oduc s, en i onmen al e ec s o animal husband y, b eeding, animal heal h, eeding, and e en human medical issues such as xeno ansplan a ion and disease modelling [1,2]. Rega ding his, he genome o a numbe o ag icul u ally impo an animal species has been o is being comple ed [3-11]. Pig is one o he mos impo an a m animals, p o id- ing abou 103,000 housand onnes o po k o mea consump ion wo ldwide in 2012 [12]. Mo eo e , pigs can be used as a model o human diseases, such as a h i is, ca dio ascula diseases, diabe es and obesi y, because pigs a e mo e simila o humans a physiological and gene le el, when compa ed wi h oden animal models [2]. Acco ding o di e en sou ces, he p edic ed numbe o pig b eeds and lines ange om 350 o 730 [13,14]. Mos o hese b eeds a e local, wi h only 25 ound in mul iple egions o a coun y, and a u he 33 sp ead o mo e han one coun y [13]. In spi e o he la ge numbe o pig b eeds, only six (La ge Whi e, Du oc, Land ace, Hampshi e, Be kshi e and Pie ain) domina e he po k indus y [13]. In he las decade, eno mous e o s ha e been made o exploi he gene ic and genomic esou ces o pigs. Genome sequencing o swine goes back o he ea ly 2000’s, when he Sino-Danish Pig Genome P ojec was ini ia ed and subsequen ly a 0.66× co e age genome su - ey, based on sho gun sequencing, was published [15]. * Co espondence: [email p o ec ed];[email p o ec ed] 1 Ag icul u al Genomics and Bioin o ma ics G oup, Ag icul u al Bio echnology Ins i u e, NARIC, Gödöllő, Hunga y Full lis o au ho in o ma ion is a ailable a he end o he a icle © 2014 Molná e al.; licensee BioMed Cen al L d. This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/2.0), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly c edi ed. Molná e al. BMC Genomics 2014, 15:761 h p://www.biomedcen al.com/1471-2164/15/761 Deepe co e age sequencing o he pig genome was ini i- a ed by he Swine Genome Sequencing Conso ium [16]. The Ssc o a9 genome assembly was eleased in 2009 [17] and he pig genome sequence was ecen ly published [9]. These genome esou ces o pig, oge he wi h specialised sequencing p ojec s such as pa allel se- quencing, ha e had a huge impac on widening ou know- ledge abou he pig genome, o include SNP iden i ica ion and geno yping [18-20], GC a iance [21], muscle an- sc ip ome [22,23], pig in e ac ome [24], domes ica ion/se- lec ion [25], e olu ion/domes ica ion [9], and in a numbe o o he ecen ly published esea ch opics [26]. Despi e he la ge numbe o local pig b eeds, only a ew o hem ( o example Angle Sa leschwein, B i ish Saddleback, Cin a Senese, Manchado de Jabugo, Basque and Guodye bas), we e included in genome sequencing p ojec s. In addi ion o he majo indus ial and he ew local b eeds, Asian and Eu opean wild boa s, se e al Asian pig b eeds and se e al o he species o he Sus genus ha e also been included [9,27-29]. Howe e , o he local b eeds, o which many a e endange ed, should also be o g ea in e es o genomic s udies because o hei impo ance in biodi e si y, conse a ion, local commu- ni y and e en po k p oduc ion issues [14,30]. Mangalica is an example o a local/ a e b eed wi h a cha ac e is ic cu ly hai pheno ype, which is indigenous o Hunga y and was de eloped in he 19 h cen u y [14]. Mangalicas a e a y- ype pigs [31], wi h high in amuscula a con- en [32]. Mangalicas ha e h ee colou a ian s, Blond, Red and Swallow-belly, which a e conside ed as sepa a e b eeds based on mic osa elli e s udies [33]. As he his- o y o he h ee Mangalica b eeds indica e [14], he Blond was b ed i s om old Hunga ian pig aces and pigs o Medi e anean o igin, and hen i con ibu ed o he wo newe b eeds, Red and Swallow-belly Mangalicas. Rep oduc ion s udies a e qui e nume ous in Mangalica [34-38], bu gene ic s udies a e a e [39]. P e iously we ha e desc ibed ha he m DNA D-loop sequences o Man- galicas display low di e si y, bu he ma e nal lineages ha hey ep esen a e gene ically dis an om cosmopoli an b eeds kep in Hunga y [14] and e y likely o igina e om one pa icula Eu opean ancien line [40]. In o de o explo e how he genomes o Mangalicas di e om he e e ence pig genome, we ha e sequenced a male indi idual o each o he h ee Mangalica b eeds along wi h a male Du oc indi idual o Hunga ian o igin. The genome sequence o Mangalicas can se e as a basis o u u e conse a ion o he b eeds and o an ex- ended Mangalica po k indus y. Resul s Genome sequencing Th ee Mangalica male pigs wi h a Mangalica-speci ic mi ochond ial D-loop haplo ype we e selec ed [40] o genome sequencing. These animals we e kep a Emőd, Hunga y, egis e ed a he Hunga ian Mangalica gene- bank as pedig ee si es. They we e p e iously assessed as Blond, Red and Swallow-belly Mangalicas, espec i ely, unde he Hunga ian Mangalica S anda d and by mic o- sa elli e analysis. A Du oc male o Hunga ian o igin was also sequenced, because we ha e ound p e iously ha Du oc pigs o in e na ional o Hunga ian o igin belong o di e en ma e nal lineages [40] and Mangalica × Du oc F1 hyb ids a e p ocessed a indus ial scale in Hunga y o po k p oduc s. Genome sequencing esul ed in 6.27 × 10 8 , 4.15 × 10 8 , 4.06 × 10 8 and 3.32 × 10 8 eads o he genomes o he Blond, Red and Swallow-belly Mangalica and he Du oc animals, espec i ely (Table 1). Due o he 500 bp a e - age agmen size o he lib a ies used o he 2 × 100 bp pai ed-end sequencing, 300 bp long space be ween he eads was p edic ed. Mapping o he eads o he e e - ence pig genome Ssc o a 10.2 esul ed in an excellen co espondence be ween he expec ed and obse ed leng h o he space s (Addi ional ile 1). The p opo ion o he mapped eads was 77.3, 83.3, 82.8 and 82.5% esul ing in 19×, 14×, 14× and 11× median au osomal co e age, espec i ely, o he ou sequenced indi iduals (Table 1). The co e age o he indi idual au osomes a ied be ween 10× and 21×, while o he sex ch omo- somes abou hal o he au osomal co e age was ob- ained (Figu e 1). In addi ion, la ge numbe s o eads o he Blond (260,270), Red (98,832) and Swallow-belly (104,478) Mangalicas and he Du oc (100,663) indi idual esul ed in 1,571×, 602×, 638× and 615× co e age o he pig e e ence mi ochond ial genome [41], espec i ely. Iden i ica ion o gene ic a ian s To iden i y SNP and INDEL a ian s we used he SAM- ools and GATK pipelines. In each animal, SAM ools and GATK p o ided a e y simila numbe o SNPs and he p opo ion o he conco dan a ia ions was high. In con as , GATK de ec ed mo e INDELs han SAM ools, and hus he p opo ion o he common INDELs was lowe compa ed ei he o he SNPs o o he o al num- be s o INDELs iden i ied by he pipelines (Addi ional ile 2). We analysed only he conco dan a ian s u he . Mo e han se en million SNP and INDEL a ian s we e iden i ied by compa ing he genome o each Mangalica indi idual o he Ssc o a 10.2 genome assembly. The genome sequence o he Du oc male also con ained al- mos 6.5 million SNPs and INDELs when compa ed wi h he e e ence genome, which was assembled p edominan ly om a Du oc emale animal [20]. SNPs ou numbe ed INDEL a ia ions in all ou animals by abou 10- old. In he Blond Mangalica, mo e homozygous hen he e ozygous SNPs we e iden i ied; in he Red Mangalica hei numbe was abou he same, while in he Swallow-belly Mangalica Molná e al. BMC Genomics 2014, 15:761 Page 2 o 12 h p://www.biomedcen al.com/1471-2164/15/761 he e we e mo e he e ozygous han homozygous SNPs. In he Du oc animal, he e we e mo e he e ozygous han homozygous SNPs. In each indi idual, mo e homo- zygous han he e ozygous INDELs we e ound and hei a io was also abou he same. SNP ansi ions we e mo e nume ous han ans e sions in all ou indi id- uals by abou 2- old. A summa y o he s a is ics o hese da a a e shown in Table 2. Fil e ing he SNP a ia ions using s ingen c i e ia (see Me hods) esul ed in 6.2 × 10 6 , 6.3 × 10 6 , 6.2 × 10 6 and 5.4 × 10 6 SNPs in he Blond, Red and Swallow-belly Mangalica and he Du oc indi iduals, espec i ely (Addi ional ile 3). App oxima ely 9 o 13% o he il- e ed SNPs we e e ealed as no el (Addi ional ile 3) when compa ed wi h he 28.6 million SNPs in he pig dbSNP138 da abase. The il e ed SNPs we e g ouped in o main and sub-ca ego ies acco ding o hei in e - genic o genic posi ion and synonymous o non- synonymous na u e (Addi ional ile 4). I was obse ed ha Mangalicas, in con as o he Du oc animal, had mo e homozygous han he e ozygous a ia ions in almos all SNP ca ego ies. A compa ison o bo h syn- onymous and non-synonymous exonic SNP a ian s e ealed 12,448 SNPs ha we e common o he ou animals, and app oxima ely 5,200 o 9,500 unique SNPs o each indi idual (Figu e 2). The de ec ion o la ge INDELs was no he scope o he cu en s udy, and so only INDELs sho e han 52 bp we e iden i ied. Fo he genomes o he Blond, Red, Swallow-belly Mangalicas and he Du oc pig, ap- p oxima ely 6.9 × 10 5 , 6.2 × 10 5 , 6.1 × 10 5 and 4.5 × 10 5 such INDELs we e iden i ied, espec i ely. O hese, 99.9% we e no el compa ed o he dbSNP138 da abase. Wi h espec o he size dis ibu ion, o he INDELs among he ou genomes, single base-pai INDELs we e he mos abundan (Addi ional ile 5). Exonic INDELs we e so ed in o eigh ca ego ies: ame-shi dele ions, ame-shi inse ions, ame-shi block subs i u ions, non- ame-shi dele ions, non- ame-shi inse ions, non- ame-shi block subs i u ions, s op-gains and s op-losses (Addi ional ile 6). In exonic INDELs, apa om he ela i ely la ge numbe o one base-pai a ia ions ha cause ORF shi s, +/−3 base-pai changes, which do no e ec heORF,we eiden i iedinhighe numbe s han wo o ou base-pai a ia ions (Addi ional ile 7). An ele- a ed numbe o one base-pai INDELs when compa ed wi h o he sizes has also been epo ed by o he s [42,43]. Ou compa ison wi h he pla inum human exonic da a ob ained om Illumina’s BaseSpace (h ps://basespace. illumina.com/da acen al) p o ided he same esul (da a no shown) sugges ing ha ou analysis wi h he pig genome is eliable. Table 1 Sequencing s a is ics BM a RM a SM a D a To al eads 626951708 414579434 405954574 331599252 Mapped eads 484893153 345426413 335911424 273390375 Mapped eads (%) 77.3 83.3 82.8 82.5 Au osomal median co e age 19 14 14 11 a BM, Blond Mangalica; RM, Red Mangalica; SM, Swallow-belly Mangalica, D, Du oc. 0 5 10 15 20 25 123456789101112131415161718XY Ch omosome Co e age (x) Blond Mangalica Red Mangalica Swallow-belly Mangalica Du oc Figu e 1 Sequence co e age o he au o- and sex-ch omosomes in ou pig indi iduals. Molná e al. BMC Genomics 2014, 15:761 Page 3 o 12 h p://www.biomedcen al.com/1471-2164/15/761 Copy numbe a ian s (CNVs) we e iden i ied ha we e common amongs he h ee sequenced Mangalicas. Only CNV gains we e analysed u he due o he e ec o sequence co e age dep h on CNV losses [44]. One housand and o y-one CNV gains wi h a copy numbe o h ee o mo e we e iden i ied ac oss all ch omosomes (Figu e 3). The minimum and maximum size o he CNVs was 1,000 and 135,735 bp, espec i ely wi h an a e age o 3,529 bp. O he 1,041 Mangalica CNVs, 485 and 160 had no posi ional o e lap wi h ei he he 3,118 CNV gains desc ibed by Paudel and colleagues [44] o he 145,857 CNVs iden i ied in he Du oc animal in his s udy, espec i ely, while he numbe s o o e lapping CNVs we e 556 and 881, espec i ely. We no e he e ha he e y la ge numbe o CNVs in he Du oc animal is because no s a is ical es could be pe o med on da a om one indi idual. Po cine genes could be anno a ed o 155 CNVs, while 886 CNVs did no con ain any gene (Addi ional ile 8). O he 155 genes, 150 we e unique since i e genes con ained wo CNVs. An o e ep esen a ion analysis iden i ied 16 ou o he 150 unique genes, which we e in he o e ep esen ed Molecula unc ion (GO:0003674) ca ego y (P alue = 1.25 × 10 −7 ). One o he 16 genes, HOXB8, encoding a homeobox p o ein, is nei he p esen in he li e a u e [44] no in he sequenced Du oc animal used in his s udy (Addi ional ile 8). Analysis o genes wi h exonic, non-synonymous SNPs Func ional, QTL and pa hway anno a ion o he genes Due o he impo ance o he Mangalica × Du oc hyb ids o he Hunga ian po k indus y, he 2,328 exonic, non-synonymous SNPs common o all h ee Mangalica b eeds bu absen om sequenced Du oc animal (Figu e 4) and he e e ence pig genome, we e selec ed o unc ional analysis. These SNPs in he coding egions o genes, which esul in amino acid changes in p o eins, may be o g ea impo ance as hey could be he polymo phisms a ec ing a ia ion in pheno ypes. The 2,328 SNPs we e mapped o 1,389 unique genes o he Ssc o a10.2 assembly as ce ain genes had mul iple SNPs (Addi ional ile 9) and hei anno- a ion in o biological p ocess (BP) ca ego ies by he web- based so wa e PANTHER [45] e ealed ha hey belong o wel e majo GO g oups (Figu e 5). Since he SNPs we e iden i ied by compa ing Mangalicas, which a e a y- ype o pigs, and Du oc, which is a lean- ype b eed, we we e pa - icula ly in e es ed in hose SNP-ha bou ing genes ha migh be in ol ed in a - ela ed biological p ocesses. Amongs he 1,389 unique genes wi h exonic, non- synonymous SNPs, we ha e iden i ied 52 genes, which belonged o Lipid me abolic p ocess (GO:0006629). Al- hough his ca ego y, in con as o when wo se s o 1,389 andomly chosen genes we e used as con ol, appea ed in an o e ep esen a ion analysis, i was no o e ep esen ed using he s ic Bon e oni co ec ion (Addi ional ile 10). As ano he con ol, we ha e ound no o e ep esen a ion using he ull pig gene se . Despi e he lack o o e ep esen- a ion, we s ill conside ha heiden i iedgenesmigh ha e a g ea impo ance, since he amino acid changes caused by he SNPs in hem may a ec he s uc u e and, conse- quen ly, he unc ion o he encoded p o ein, and such unc ional al e a ions o p o eins emain hidden in gene ex- p ession s udies. The impo ance o ou SNP-based gene iden i ica ion app oach is indica ed by, o example, ha p o eins encoded by he PNLIP and PNLIPRP2 genes, which we e no associa ed o a ness pheno ypes in pigs be- o e, a e he a ge o O lis a ( e ahyd olips a in), a d ug used o ea ing obesi y in humans (da a no shown). The possible e ec o exonic SNPs on p o ein unc ion is dis- cussed below using FASN as an example. To s udy he possible ela ionship be ween he 52 genes in he lipid me abolic p ocess GO ca ego y and QTLs, he ch omosomal posi ion o each genes was compa ed o he Figu e 2 Venn diag am o exonic SNPs in he sequenced animals. D, Du oc; BM, Blond Mangalica; SM, Swallow-belly Mangalica; RM, Red Mangalica. Table 2 Ca ego ies o sequence a ia ions BM a RM a SM a D a SNPs 6944767 6871283 6734038 5950027 INDELs 696029 623282 617600 451299 To al a ian s 7640796 7494565 7351638 6401326 He e ozygous SNPs 3196568 3443594 3722921 3314982 Homozygous SNPs 3744651 3424501 3008107 2633108 SNP ansi ions 4782645 4739843 4641040 4097573 SNP ans e sions 2165670 2134628 2096008 1854391 Mul iple SNPs 3548 3188 3010 1937 He e ozygous INDELs 210860 200062 167240 144505 Homozygous INDELs 464812 406643 436692 298970 a BM, Blond Mangalica; RM, Red Mangalica; SM, Swallow-belly Mangalica, D, Du oc. Molná e al. BMC Genomics 2014, 15:761 Page 4 o 12 h p://www.biomedcen al.com/1471-2164/15/761 posi ions o he “Fa ness”and “Fa composi ion”QTLs downloaded om he QTLdb, Release 19, [46]. Fo y-nine genes a e in one o mo e a - ela ed QTLs wi h 14 genes on ch omosome 14, o e lapped by 15 a -associa ed QTLs (Addi ional ile 11). Because o his la ge p opo ion (~28%) o genes on ch omosome 14, we pe o med an en ichmen analysis o he 14-gene se and a con ol se o 1282 genes, bo h a e in he same egion o ch omosome 14 de e mined by he 15 QTLs. The co ec ed P- alue o lipid me abolic genes in he con ol and in ou se was 4.80 × 10 −3 and 2.95 × 10 −19 , espec i ely, indica ing ha he en ichmen o he 14 genes in hese QTLs de ia e signi ican ly om andom. Fa y acid composi ion o mea s is an impo an die - e ic and heal h issue o po k consume s. We, he e o e, compa ed hose genes, which a e in sa u a ed and unsa - u a ed a y acid QTLs and ound ha nine genes we e in common ac oss bo h a y acid ca ego ies, while he sa u a ed and unsa u a ed QTL g oups each con ained wo unique genes, NKX2-3 and EPHX2, and OMA1 and FAM135B, espec i ely (Addi ional ile 12). Figu e 3 Dis ibu ion o CNVs ac oss Mangalica ch omosomes. Sho e ical lines ep esen he posi ion o CNVs, which a e p esen in all h ee sequenced Mangalicas. Figu e 4 Venn diag am o exonic, non-synonymous SNPs in he sequenced animals. D, Du oc; BM, Blond Mangalica; SM, Swallow-belly Mangalica; RM, Red Mangalica. Molná e al. BMC Genomics 2014, 15:761 Page 5 o 12 h p://www.biomedcen al.com/1471-2164/15/761 O he 52 lipid me abolic p ocess-associa ed genes, we could map 41 o one o mo e pa hways using he KEGG da abase. Almos 44% (18) o he mapped genes we e associa ed wi h lipid me abolic pa hways (Figu e 6), while o he s con ibu e o glycan and ca bohyd a e me- abolisms, biochemical p ocesses a he in e ace o lipid and o he me abolic pa hways and he egula ion o lipid me abolism (Addi ional ile 11). O he 41 mapped genes, wo a e pa icula ly impo an . One is FASN, which encodes an enzyme in ol ed in a numbe o s eps in he syn hesis o 8 o 16 ca bon-chain a y acids in he a y acids biosyn hesis pa hway [KEGG:ssc00061]. The FASN p o ein is a homodime ic mul i unc ional enzyme wi h six ca aly ic domains, which p ocesses di - e en s eps o cyclic elonga ion o a y acids [47]. The o he gene is SLC27A6, a membe o a gene amily, which is exp essed in li e , hea and subcu aneous back a o pig [48]. The encoded p o ein is a a y acid anspo e , which is one o he wo memb ane p o eins o he PPAR signalling cascade [KEGG:hsa03320], which egula e lipid and a y acid me abolism, bile acid bio- syn hesis and adipocy e di e en ia ion, amongs o he egula ed p ocesses [49]. Geno yping SNPs in o he b eeds The 90 SNPs in he abo e desc ibed 52 genes we e p esen in all h ee sequenced Mangalicas, bu absen om he sequenced Du oc and he e e ence genome. To lea n abou hei wide occu ence, we ha e “e-geno yped” 55 animals whose genome was sequenced [9] o hese SNPs. The esul s indica e ha he equencies o hese SNPs a y amongs he 55 indi iduals (Addi ional ile 13). Clus e ing o he a e age equencies e ealed ou clus- e s among he indi iduals, whe e Mangalica ep esen s a sepa a e clus e and Eu opean, in e na ional/Hunga ian Du oc, and non-Eu opean pigs and/o wild boa s com- p ise he h ee o he ela ed g oups (Addi ional ile 14). The clea sepa a ion o Mangalicas om o he b eeds by hose 90 SNPs migh ha e he po en ial in p ac ical appli- ca ions, such as whole genome selec ion in b eeding. I was ound ha ou SNPs a e p esen only in Man- galicas, bu no in he geno yped indi iduals (Addi onal ile 13). All o hese SNPs a e in one gene, MOGAT2 (ENSSSCG00000014861), which encodes a monoacyl- glyce ol O-acyl ans e ase 2 enzyme, and is in se e al back- and belly- a QTLs and in he “Fa diges ion and abso p ion”(KEGG: 04975) pa hway (Addi ional ile 11). I is possible, he e o e, ha his gene has a pa icu- la ole o he de elopmen o he a y-pig pheno ype o Mangalicas. Some s udies ha e highligh ed he impo ance o he FASN gene in pig a ness [50,51]. In his gene, we ha e iden i ied wo non-synonymous SNPs, which a e p esen in he h ee sequenced Mangalicas, bu no in he e e - ence genome and he sequenced Du oc indi idual used in his p ojec . They a e also di e en om hose h ee Figu e 5 Biological p ocess on ology o genes wi h exonic SNPs ound in Mangalica b eeds. O he 1.389 genes, 1,372 esul ed in 2,130 o al hi s in p ocesses. Pe cen age indica es he pe cen o genes in one p ocess agains he o al numbe o p ocess hi s. Molná e al. BMC Genomics 2014, 15:761 Page 6 o 12 h p://www.biomedcen al.com/1471-2164/15/761 SNPs ha ha e been geno yped p e iously [50]. SNP1 is in exon 9 (ch omosome 12, posi ion 1,028,766) and is a G•C ( e e ence) o A•T (Mangalica) ansi ion, which causes a R443Q amino acid change while SNP2 is a C•G ( e e ence) o T•A (Mangalica) ansi ion in exon21 (ch omosome 12, posi ion 1,025,096) esul ing in a T1088I change in he FASN p o ein. The equency o hese wo SNPs is qui e di e se in he genome sequenced animals, in- cluding he h ee Mangalicas and one Du oc indi idual se- quenced in his s udy (Addi ional ile 15). We, he e o e, geno yped 72 Mangalica and 21 Du oc pigs o bo h SNPs in o de o ge mo e in o ma ion abou hese SNPs in he wo b eeds. We ound ha he A (“Mangalica”) alleles (SNP1 A•T o SNP2 T•A ) occu s a a much highe equency han he B alleles in Mangalica, whe eas in con as he B alleles (SNP1 G•C o SNP2 C•G ,“non-Mangalica”)a emo e p e alen in Du oc (Table 3). Addi ionally, we ound ha o SNP1, 62 and 10 Mangalicas and 1 and 20 Du oc ani- mals we e AA and BB homozygous, espec i ely; no he e o- zygo es we e ound. Fo SNP2, 65 Mangalicas and eigh Du ocs had AA, i e Du ocs had BB, and se en Mangalicas and eigh Du ocs had AB geno ypes espec i ely; no Mangalica wi h BB geno ype was ound. Discussion The genome o one indi idual each o he h ee Mangalica b eeds (Blond, Red and Swallow-belly), and a Du oc animal om a Hunga ian he d was sequenced and analysed. Mo e han 100 million eads we e ob ained om he genome o each animal. On a e age o he ou genomes sequenced, 81% o he eads we e mapped o he e e ence genome, esul ing in 14.5× median au osomal co e age. Millions o SNP and hund ed- housands o INDEL a ia ions we e iden i ied in he h ee Mangalicas and he one Du oc gen- ome, espec i ely, when compa ed o he e e ence pig gen- ome assembly Ssc o a 10.2. By il e ing he SNPs, abou i e o six million a ia ions we e ob ained, and abou one- en h o hese we e no el SNPs compa ed o he dbSNP138 da abase (Addi ional ile 3). Fo unc ional analysis, we selec ed 2,328 exonic non- synonymous SNPs p esen in each sequenced Mangalica Table 3 Geno yping he FASN gene B eed Allele (Nucleo ide) Allele equency SNP1 Mangalica A (A•T) 0.86 Mangalica B (G•C) 0.14 Du oc A (A•T) 0.05 Du oc B (G•C) 0.95 SNP2 Mangalica A (T•A) 0.95 Mangalica B (C•G) 0.05 Du oc A (T•A) 0.05 Du oc B (C•G) 0.95 Fay acid me abolism PLA2G4F,PTGIS, EPHX2, ACACA PLA2G3, PLA2G1B, ALDH3A2, ACAA Fay acid biosyn hesis ACOT4, BAAT, ACACA, ACAA, FASN Fay acid elongaon ACOT4 Glyce olipid me abolism GPAT2, PNLIP, PNLIPRP2 Fa digeson and abso pon PLA2G3, PNLIP, MOGAT2, MTTP PLA2G1B, PNLIPRP2, GPAT2 Glyce ophospholipid me abolism PLA2G4F,GPAT2, PLA2G3 Bile acid biosyn hesis CYP46A1,BAAT Bile sec eon BAAT Panc eac sec eon PLA2G3, PNLIP, PLA2G1B, PNLIPRP2 Die a y a PPAR signalling pa hway ACAA,SLC27A6 Chylomic on Fay acids Figu e 6 Fa me abolic pa hways and pa icipa ing genes wi h Mangalica-speci ic exonic, non-synonymous SNPs. Lines ep esen he in e connec ions o he pa hways. A ows indica e whe e signalling o me aboli es (name abo e he line) a ec genes in o he pa hways. Molná e al. BMC Genomics 2014, 15:761 Page 7 o 12 h p://www.biomedcen al.com/1471-2164/15/761 indi idual, bu absen om ei he he e e ence genome o he Hunga ian Du oc animal. These SNPs we e mapped o 1,389 pig genes p esen in he Ensembl da abase. Since Mangalicas a e a y- ype pigs, and he SNPs we e iden i ied in compa ison wi h Du oc, a lean- ype pig, we we e pa - icula ly in e es ed in a - ela ed genes in his se . Fi y- wo genes we e ound belonging o lipid- ela ed me abolic p ocess ca ego ies and we e u he analysed using QTL and pa hway da a-mining. O he 52 genes, 49 and 41 a e associa ed wi h a - ela ed QTL egions and KEGG pa h- ways, espec i ely (Addi ional ile 11). Some o he 52 genes, o example ACACA,ANKRD23, GM2A,KIT,MOGAT2,MTTP,FASN,SGMS1,SLC27A6 and RETSAT, which we ha e highligh ed he e, ha e been p e iously desc ibed in he con ex o a - ela ed cha ac e - is ics in pigs [50-54]. O hese genes, FASN, a gene encod- ing a a y acid syn hase, has been shown o be associa ed wi h a cis-11-Eicosenoic acid (C20:1) pe cen age QTL in a Guadye bas × Land ace c oss, al hough none o he iden i- ied SNPs had any pu a i e e ec on he p o ein s uc u e [50]. The FASN p o ein is a homodime ic, mul i unc ional enzyme wi h six ca aly ic domains, which a e equi ed o he cyclic elonga ion o a y acids [47] and ca alyses 32 e- ac ions in he a y acid biosyn hesis [KEGG:ssc00061] pa hway. Ta ge ed mu agenesis o he FASN gene and in- hibi ion o he FASN p o ein in mice esul ed in educed o al body a [55] and body weigh [56], espec i ely. We ha e iden i ied wo SNPs in his gene in Mangalicas ha e- sul in a R443Q (SNP1) and a T1088I (SNP2) amino acid change. The amino acid in posi ion 443 is pa o he α- helix in he p o ein’s in e -domain linke . Since glu amine is mo e hyd ophilic han a ginine, he amino acid subs i u- ion may a ec he ela i e posi ion o he wo unc ional domains by modula ing he lexibili y o he linke connec - ing hem [57]. The amino acid in posi ion 1,088 is pa o he dehyd a ase domain o he FASN p o ein. This domain ca alyses he con e sion o β-hyd oxyacyl-ACP o β-enoyl- ACP in he cyclic elonga ion o a y acids [47]. T 1088 is in close icini y o he ac i e si e o he dehyd a ase domain con aining an open-ended hyd ophobic unnel [57]. P e- dic ing hyd ophobici y o amino acids along he FASN polypep ide e ealed ha he subs i u ing I 1088 is s ongly hyd ophobic, while T 1088 is hyd ophilic (da a no shown). I is possible, he e o e, ha in he FASN T1088I p o ein he subs a e-binding na u e o he ac i e si e is al e ed, which may in luence he dehyd a ion s ep o he a y acid cyclic elonga ion. This migh be pa icula ly impo an in Mangalicas, whe e no BB homozygo es we e ound. Thus he ac i e si e in he ca aly ic domain o hei FASN p o ein is expec ed o be hyd ophobic, al hough allele-speci ic exp ession o he FASN gene in he e o- zygo es migh in luence his. I is known ha eeding egimes in luence a y acid composi ion and mea ’s ma bling in Mangalicas [31,58], simila o o he pig b eeds and a m animals. In lipid me abolism, he “Fa diges ion and abso p ion”and “Bile sec e ion”pa hways a e in ol ed in he me abolism o die a y a s. These wo pa hways a e connec ed o he “Glyce olipid me abolism”,“Fa y acid me abolism”and “Fa y acid biosyn hesis”pa hways. Ou s udy highligh ed a numbe o genes in hese me abolic pa hways and in he PPAR signalling pa hway (Figu e 6). We ha e iden i ied one gene, MOGAT2 (ENSSSCG00000014861), wi h se en SNPs, o which ou a e p esen in Mangalicas, bu no in o he 56 sequenced pig indi iduals (see Resul s). The MOGAT2 p o ein ca alyses he con e sion o 1-acylglyce ol ob ained om die a y a in o diacylglyc- e ol in he smoo h endoplasma ic e iculum o he small in es inal epi helial cells, and hus pa icipa es in he p oduc ion o chylomic on (“Fa diges ion and abso p ion”pa hway, KEGG:04975). Chylomic on a - ec s he PPAR signalling pa hway, which in u n egula es a numbe o lipid me abolic p ocesses (Figu e 6). I is possible, he e o e, ha polymo - phisms ha a ec genes in his complex ne wo ks o pa hways, which a e also pa o ele an QTLs, may be esponsible o he di e ences in a ening, a com- posi ionandany ela edpheno ypes ha we eob- se ed be ween b eeds in esponse o di e en eeding egimes. Fo example, he MOGAT2 gene was ound o be pa o he lipid concen a ion biological unc ion, modula ed in back a [54]. Conclusions The disco e y o genes behind ag icul u ally impo an ai s is a di icul ask in a m animals, in pa icula when he in e media e- o end-pheno ypes a e de e - mined by QTLs. In his s udy, we desc ibed he gen- ome sequencing and analysis o h ee Hunga ian Mangalica indi iduals ep esen ing each o he h ee Mangalica b eeds, which a e local, a y ype pigs wi h a niche ole in he po k ma ke . A e il e ing, millions o SNPs we e iden i ied in each animal compa ed o he e e ence genome, and abou 10% o hem a e no el compa ed o he po cine SNP en ies o he dbSNP138 da abase. This inding highligh s ha sequencing genomes o indi iduals o a e/local b eeds can p o ide la ge amoun s o da a iden i ying genomic a ia ions ela i e o he e e ence genome o he same species. These a ia ions can be he basis o gene disco e ies. Wi h special emphasis on pig a ness, by anno a ing and compa ing exonic, non- synonymous Mangalica-speci ic SNPs o QTLs and pa h- ways, we iden i ied a numbe o candida e genes, which can se e o u u e geno yping, exp ession, s uc u e- unc ion, and biological ne wo k s udies and in applica ions, such as molecula b eeding and mea iden i ica ion o acing in bo h Mangalica and o he b eeds. Molná e al. BMC Genomics 2014, 15:761 Page 8 o 12 h p://www.biomedcen al.com/1471-2164/15/761 Me hods Genome sequencing Pig blood samples we e ob ained om he MAN- GFOOD conso ium’s Biobank a he Ag icul u al Bio- echnology Cen e , Gödöllő, Hunga y. To al DNA was ex ac ed using he Duplicα® P ep Au oma ic Ex ac ion Sys em and he Duplicα® Blood DNA ki (Eu oClone, Milan, I aly). DNA concen a ion was measu ed using he Quan -iT™PicoG een dsDNA® Assay (Li e Technolo- gies, Budapes , Hunga y). P epa a ion o 500 bp agmen lib a ies and 2 × 100 bp Illumina pai ed-end genome se- quencing was pe o med by A os Applied Bio echnology (Aa hus, Denma k) as a cus om se ice, using Illumina’s HiSeq2000 pla o m. Da a analyses The Sus sc o a e e ence genome sequence 10.2 was indexed using he “bw sw”algo i hm op ion o BWA 0.5.9 c1 [59] ollowed by mapping he sho sequence eads o he indexed genome using he de aul se ings and he pai ed-end me hod o he same so wa e. The ob ained BAM iles we e so ed and indexed o u he analyses. To de ec small gene ic a ian s (SNPs and INDELs), he SAM ools [60] and GATK ( e sion: 2.3-9-ge5eb 34) [61] a ian calling pipelines we e employed. In SAM- ools, base-calling was pe o med using he “mpileup” command and he “-E -D -S -u”pa ame e s o SAM ools 0.1.18. The “ iew”command o BCF ools was used o call he a ian s using he “-b cg”pa ame e s. VCF iles we e hen gene a ed by he “ c u ils.pl”sc ip using he “ a Fil e ”op ion and SNPs and INDELs we e ex ac ed. Finally, SNPs, which had a Ph ed sco e highe han 30 (i.e. hei base-calling accu acy is la ge hen 99.9%), and a high-quali y ead co e age o minimum h ee, we e il e ed using a cus om sc ip . INDELs we e used in downs eam analyses wi hou il e ing. Fo GATK, he dbSNP138 da a we e used as a aining se . O he se ings we e used ac- co ding o he GATK bes p ac ice online documen a ion. Resul s ob ained by he wo pipelines we e compa ed using he BEDTools’[62] “in e sec Bed”module o SNPs and using ou cus om sc ip o INDELS; only conco dan a i- a ions we e p ocessed u he . Copy numbe a ia ions (CNVs) we e de ec ed as de- sc ibed by Paudel and cowo ke s [44] using he m Ca- NaVa ( e sion 0.51) so wa e [63]. The window size was se o 1,000 bp. We selec ed windows whe e he copy numbe and he s anda d de ia ion we e bigge han h ee and 0.7, espec i ely, o he h ee Mangalicas. A e ha s ep he egions we e chained. To de e mine no el a ian s in ou sequence da a, we compa ed he iden i ied SNPs and INDELs wi h he dbSNP138 da a using BEDTools [62] and anno a ed he de- ec ed gene ic a ian s using ANNOVAR [64]. Following he ANNOVAR analysis, non-synonymous exonic SNPs, which we e p esen only in Mangalicas, we e de e mined by BEDTools’“mul iIn e sec Bed”module. Genes ca ying hese a ian s we e iden i ied using a cus om sc ip . Com- pa ison o SNPs in he lipid me abolism genes amongs genome sequenced animals ( his s udy and li e a u e 44) we e also pe o med using he “mul iIn e sec Bed”module o BEDTools. Gene on ology analysis was pe o med by he web- based so wa e PANTHER [45]. Fo o e ep esen a ion analyses, Bioma ’s [65] en ichmen analysis op ion wi h 0.05 cu o P- alue was employed using he Ssc o a 10.2 e e ence genome as backg ound. Random se s o genes was gene a ed by a cus om Py hon sc ip . Fa - ela ed pig QTLs and hei posi ions we e downloaded om he QTLdb (Release 19) da abase [46], and hei ex ension was compa ed wi h he posi ion o he SNPs o selec ed genes manually. Genes we e anno a ed in o pa hways using he KEGG da abase. Da a om Ensembl we e e ie ed using BioMa [65]; Venn diag ams we e gene a ed using he so wa e Venny [66]; clus e ing was pe o med using CIMmine [67] wi h Manha an dis ance and comple e linkage clus e - ing se ings. Geno yping To geno ype he wo Mangalica-speci ic SNPs in he FASN gene, High Resolu ion Mel ing (HRM) analysis was pe - o med wi h a Ro o -Gene Q 5plex HRM Pla o m using a sa u a ing dye (E aG een) echnology (Qiagen, Hilden, Ge many). PCR eac ions we e pe o med in 25 μl eac ion olumes using 60 ng o al DNA as empla e and he Type- i HRM PCR ki (Qiagen, Hilden, Ge many), acco ding o he ins uc ion o he manu ac u e . The p ime s o FASN SNP1 and SNP2 we e FASN1_F: 5′CGCGATCTCGTT GAGCAT 3′,FASN1_R:5′GTGCAGACCCTGCTGGAG 3′and FASN2_F: 5′GGATAGGCTTGAGATGCTCTT 3′, FASN2_R 5′GTGGTGGTGGACAGGAATCT 3′, e- spec i ely. Reac ions we e ca ied ou wi h an ini ial de- na u a ion s ep a 95°C o 5 min, ollowed by 35 cycles o 95°C o 15 sec, 60°C o 30 sec and 72°C o 10 sec and hen HRM cu es we e gene a ed by acqui ing lo escence da a be ween 80 and 91°C. Indi iduals wi h homozygous and he e ozygous geno ypes we e assigned acco ding o hei HRM cu e de e mined by he Ro o -Gene so wa e and isual inspec ion. A ailabili y o suppo ing da a The da a se s suppo ing he esul s o his a icle a e included wi hin he a icle and i s addi ional iles. Se- quence da a a e deposi ed o he NCBI Sequence Read A chi e unde iden i ie SRP039012. Molná e al. BMC Genomics 2014, 15:761 Page 9 o 12 h p://www.biomedcen al.com/1471-2164/15/761