ORIGINAL RESEARCH
published: 22 Feb ua y 2018
doi: 10.3389/ gene.2018.00046
F on ie s in Gene ics | www. on ie sin.o g 1Feb ua y 2018 | Volume 9 | A icle 46
Edi ed by:
Rod igo A. To es,
Uni e sidade Fede al de Pe nambuco,
B azil
Re iewed by:
Ped o Manoel Gale i J ,
Fede al Uni e si y o São Ca los, B azil
Ma ia Raquel Mou a Coimb a,
Fede al Ru al Uni e si y o
Pe nambuco, B azil
*Co espondence:
Diogo T. Hashimo o
[email p o ec ed]
Special y sec ion:
This a icle was submi ed o
E olu iona y and Popula ion Gene ics,
a sec ion o he jou nal
F on ie s in Gene ics
Recei ed: 01 Sep embe 2017
Accep ed: 31 Janua y 2018
Published: 22 Feb ua y 2018
Ci a ion:
Jo ge PH, Mas ochi ico-Filho VA,
Ha a ME, Mendes NJ, A iede RB,
F ei as MV, Ve a M, Po o-Fo es i F
and Hashimo o DT (2018) Gene ic
Cha ac e iza ion o he Fish Pia ac us
b achypomus by Mic osa elli es
De i ed om T ansc ip ome
Sequencing. F on . Gene . 9:46.
doi: 10.3389/ gene.2018.00046
Gene ic Cha ac e iza ion o he Fish
Pia ac us b achypomus by
Mic osa elli es De i ed om
T ansc ip ome Sequencing
Paulo H. Jo ge1, Vi o A. Mas ochi ico-Filho1, Milene E. Ha a1, Na ália J. Mendes1,
Raquel B. A iede1, Milena Viei a de F ei as1, Manuel Ve a2, Fábio Po o-Fo es i3and
Diogo T. Hashimo o1*
1Aquacul u e Cen e o Uni e sidade Es adual Paulis a Júlio de Mesqui a Filho, São Paulo S a e Uni e si y, Jabo icabal,
B azil, 2Ve e ina y Facul y, Uni e si y o San iago de Compos ela, Lugo, Spain, 3School o Sciences, São Paulo S a e
Uni e si y, Bau u, B azil
The pi api inga, Pia ac us b achypomus (Cha aci o mes, Se asalmidae), is a ish om
he Amazon basin and is conside ed o be one o he main na i e species used in
aquacul u e p oduc ion in Sou h Ame ica. The objec i es o his s udy we e: (1) o
pe o m li e ansc ip ome sequencing o pi api inga h ough NGS and hen alida e
a se o mic osa elli e ma ke s o his species; and (2) o use polymo phic mic osa elli es
o analysis o gene ic a iabili y in a med s ocks. The ansc ip ome sequencing was
ca ied ou h ough he Roche/454 echnology, which esul ed in 3,696 non- edundan
con igs. O his o al, 2,568 con igs had simila i y in he non- edundan (n ) p o ein
da abase (Genbank) and 2,075 sequences we e cha ac e ized in he ca ego ies o Gene
On ology (GO). A e he alida ion p ocess o 30 mic osa elli e loci, eigh ma ke s showed
polymo phism. The analysis o hese polymo phic ma ke s in a med s ocks e ealed
ha ish a ms om No h B azil had a highe gene ic di e si y han ish a ms om
Sou heas B azil. AMOVA demons a ed ha he highes p opo ion o a ia ion was
p esen ed wi hin he popula ions. Howe e , when compa ing di e en g oups (1: Wild; 2:
No h ish a ms; 3: Sou heas ish a ms), a conside able a ia ion be ween he g oups
was obse ed. The FST alues showed he occu ence o gene ic s uc u e among he
b oods ocks om di e en egions o B azil. The ansc ip ome sequencing in pi api inga
p o ided impo an gene ic esou ces o biological s udies in his non-model species,
and mic osa elli e da a can be used as he amewo k o he gene ic managemen
o b eeding s ocks in B azil, which migh p o ide a basis o a gene ic p e-b eeding
p og amme.
Keywo ds: aquacul u e, gene ic s uc u e, NGS, Pi api inga, Se asalmidae
INTRODUCTION
The pi api inga (Pia ac us b achypomus) is a na i e ish om he Amazon and O inoco Ri e s and
can each up o 20 kg o weigh (Alcân a a e al., 1990). This species is used o ish a ming, is
alued o i s mea and has as g ow h pe o mance (F esneda e al., 2004). In B azil, pi api inga
a ming ep esen s he hi d la ges ish p oduc ion ope a ion (abou 10,000 ons) among he na i e
Jo ge e al. Gene ic Resou ces in Pi api inga
ish species (MPA, 2013a). Fu he mo e, his species has
been widely used o he p oduc ion o in e speci ic hyb ids,
pa icula ly he amba inga ( emale ambaqui Colossoma
mac opomum ×male pi api inga P. b achypomus), and pa inga
( emale pacu Pia ac us mesopo amicus ×male pi api inga P.
b achypomus; IBGE, 2016). The aquacul u e p oduc ion o
pi api inga in B azil is concen a ed mainly in he Midwes and
No h (87%), ollowed by he Sou heas (9%), No heas (3%)
and Sou h (1%) (MPA, 2013b). This species also has economic
impo ance o aquacul u e in o he coun ies in Sou h Ame ica
(Colombia, Pe u, and Venezuela) and in Asia (China, Myanma ,
Thailand, and Vie nam; Flo es Na a, 2007; Honglang, 2007; Lin
e al., 2015).
Howe e , despi e his ep esen a ion o aquacul u e
p oduc ion, ew scien i ic s udies ha e ocused on unde s anding
he biology o pi api inga, especially o gene ic ai s. So, he
gene a ion o gene ic esou ces o his species is undamen al
o ad ancing s udies o b eeding and gene ic managemen , as
occu ed in model species used in aquacul u e, such as salmon,
ca ish, ca p, and ilapia (Lien e al., 2011; Liu e al., 2011; Guyon
e al., 2012; Ji e al., 2012).
Model ish species, such as zeb a ish Danio e io, ha e been
desc ibed wi h mo e han 26,000 genes (Howe e al., 2013).
Howe e , ew genes and hei me abolic pa hways ha e been
cha ac e ized o non-model species wi hou e e ence genomes,
as is he case o pi api inga. In he ield o gene ics and molecula
biology, Nex -Gene a ion Sequencing (NGS) echnologies a e
causing a e olu ion, allowing he sequencing o genome and
ansc ip ome o any o ganism, quickly and a low cos (Seeb
e al., 2011). RNA-seq ( ansc ip ome sequencing) is conside ed
one o he mos used s a egies o NGS echnology o he
ansc ip s analysis (Qian e al., 2014), whe ein all he messenge
RNA (mRNA) o a speci ic issue o se o issues a e used as a
sou ce o sequencing. Mo eo e , RNA-seq is an e ec i e ool o
disco e y o molecula ma ke s, pa icula ly o he p ospec ion
o gene-associa ed mic osa elli es (Teache e al., 2012; Xu e al.,
2013).
Due o he use ulness o e ealing he gene ic a ia ion among
indi iduals (Liu and Co des, 2004), mic osa elli e ma ke s ha e
p o en o be e icien o gene ic cha ac e iza ion o wild
popula ions and b eeding s ocks o a med ish (Koljonen e al.,
2002; Lehoczky e al., 2005), such as o p e en inb eeding
(Ponzoni e al., 2008), o iden i y and p ese e li e gene banks
(Machado-Schia ino e al., 2007), o de ec gene ic s uc u e
(Do P ado e al., 2018), o di ec ma ings du ing he o ma ion
o he popula ion base o b eeding p og ammes (Fe nández
e al., 2014), and o pe o m ma ke assis ed selec ion (MAS) o
economic ai s (Hous on e al., 2010). Howe e , hese ma ke s
a e no a ailable o pi api inga, one o mos impo an species
o he aquacul u e in Sou h Ame ica.
Fo he aquacul u e o pi api inga, analysis o gene ic
a iabili y in a med s ocks s ill needs o be pe o med, which
will allow h ee hypo heses o be es ed: (1) a med s ocks
o pi api inga ha e lowe gene ic di e si y in ela ion o wild
s ocks; (2) a med s ocks o pi api inga in B azil a e gene ically
s uc u ed; and (3) gene-linked mic osa elli es can be associa ed
o economic ai s o pi api inga, such as g ow h and disease
esis ance. These analyses will suppo he c ea ion o a b eeding
p og amme o inc ease he p oduc i i y o pi api inga, by
di ec ed ma ings which lead o he o ma ion o amilies,
a oiding he p oblems o bo lenecks and inb eeding in he base
popula ion (Fe nández e al., 2014), and by he iden i ica ion o
quan i a i e ai loci (QTL), which will assis he selec ion o
supe io geno ypes by MAS (Hous on e al., 2010).
Thus, he objec i e o he p esen s udy was o cha ac e ize
gene ic esou ces o he p ope managemen o his non-model
species in aquacul u e, h ough ansc ip ome cha ac e iza ion
and gene ic a iabili y analysis o s ocks using mic osa elli e
ma ke s.
MATERIALS AND METHODS
E hics S a emen
This s udy was ca ied ou in s ic acco dance wi h he
animal wel a e guidelines o he Na ional Council o Con ol
o Animal Expe imen a ion (B azilian Minis y o Science,
Technology, and Inno a ion). The p esen s udy was pe o med
unde au ho iza ion N◦33435-1, issued h ough ICMBio (Chico
Mendes Ins i u e o he Conse a ion o Biodi e si y, B azilian
Minis y o En i onmen ). No animal was housed o ca ed
o in he labo a o y. Fish we e eu hanized by benzocaine
anes he ic o e dose o collec ion o li e issue o ansc ip ome
sequencing. Fo mic osa elli e alida ion and gene ic a iabili y
analysis, in agmen s we e collec ed om each ish unde
benzocaine anes hesia and all e o s we e made o minimize
su e ing.
Samples o T ansc ip ome Sequencing
To pe o m he ansc ip ome sequencing, samples o li e
issue we e aken om 10 indi idual ish om h ee di e en
B azilian ish a ms and one wild popula ion: Aquacul u e
Cen e o São Paulo S a e Uni e si y, CAUNESP, Jabo icabal, SP
(n=3); P oje o Peixe ish a m, Sales Oli ei a, SP (n=1);
Fazenda São Paulo ish a m, B ejinho de Naza é, TO (n=5);
and Tocan ins Ri e , Lajeado, TO (n=1). Indi iduals om
di e en o igins we e used in o de o achie e he highes gene ic
a iabili y in mic osa elli e disco e y analysis. Li e samples
we e selec ed o ansc ip ome s udies because he li e plays
a c i ical ole in coo dina ing a ious physiological p ocesses,
including diges ion, me abolism, de oxi ica ion, and endoc ine
sys em immune esponse (Ma in e al., 2010).
Samples o Gene ic Va iabili y Analysis
Analyses o mic osa elli e alida ion we e pe o med in 22
indi idual pi api inga collec ed om he Tocan ins Ri e (TO)
om Lajeado Ci y, Tocan ins S a e, B azil. We hen used he
mic osa elli e ma ke s o s udy he gene ic a iabili y in samples
collec ed om ou comme cial ish a ms: TO1 (n=25) and
TO2 (n=26), om Tocan ins S a e (No h B azil); and SP1
(n=36) and SP2 (n=20), om São Paulo S a e (Sou heas
B azil). To main ain he con iden iali y o hese ish a ms, he
names o and in o ma ion on he ish a ms ha e been p ese ed.
F on ie s in Gene ics | www. on ie sin.o g 2Feb ua y 2018 | Volume 9 | A icle 46
Jo ge e al. Gene ic Resou ces in Pi api inga
Analysis o Gene ic Pu i y in Pi api inga
Indi iduals
Acco ding Hashimo o e al. (2014), in e speci ic hyb ids ha e
been de ec ed in b oods ocks o B azilian ish a ms. The
pi api inga can be c ossed wi h ambaqui C. mac opomum o
pacu P. mesopo amicus, esul ing in iable and e ile hyb ids
(Hashimo o e al., 2012, 2014). The e o e, in he p esen s udy,
special a en ion was gi en o analyze pu e pi api inga, and no
in e speci ic hyb ids. The analysis o gene ic pu i y in all animals
he ein s udied was pe o med using he mi ochond ial genes,
Cy och ome C Oxidase subuni I (m -co1) and Cy och ome b
(m -cyb); and he nuclea genes, α-T opomyosin ( pm1) and
Recombina ion Ac i a ing Gene 2 ( ag2), acco ding o he
p o ocols and me hods o Hashimo o e al. (2011). Fish iden i ied
as in e speci ic hyb ids we e excluded om u he analysis in
his s udy.
cDNA Lib a y Cons uc ion and Roche 454
Pla o m Sequencing
Samples o ∼100 mg o li e ixed in RNAla e we e ex ac ed
wi h Rneasy Mini Ki (Qiagen). Each sample was quan i ied
by spec opho ome y using NanoD op ND-1000 equipmen
and he quali y (in eg i y) was checked by 2100 Bioanalyze
equipmen . I succeeded he p epa a ion o an equimola pool o
o al RNA samples ( om 10 indi iduals) o mRNA en ichmen
wi h µMACS mRNA Isola ion Ki (Mil enyi Bio ech).
A non-no malized cDNA lib a y was p epa ed using cDNA
Syn hesis Sys em Ki wi h andom p ime GS Rapid Lib a y
P ep Ki and GS Rapid Lib a y MID Adap o s Ki (Roche).
The High Sensi i i y DNA LabChip Ki (Agilen Technologies)
wi h 2100 Bioanalyze was used o quali y analysis o he
cDNA lib a y. The concen a ion o sample (molecules/µL)
was ob ained by Quan iFluo TM—ST luo ime e (P omega).
Ti a ion o emPCR (emulsion PCR) was pe o med wi h he
GS FLX Ti anium SV em PCR Ki (Lib-L) (Roche), acco ding
o he emPCR Ampli ica ion Me hod Manual—Libl SV, GS
FLX+Se ies, o iden i y he op imal numbe o DNA molecules
pe bead (cpb =copies pe bead). A e emPCR i a ion, he
emPCR was pe o med wi h GS FLX Ti anium LV emPCR
Ki (Lib-L) (Roche), acco ding o he emPCR Ampli ica ion
Me hod Manual—LibL LV, GS FLX+Se ies. The ansc ip ome
sequencing was conduc ed using he Roche/454 echnology (GS
FLX Ti anium Sequencing Ki XL +) om HELIXXA company
(Campinas, SP, B azil), which has been used o ansc ip ome
analysis o non-model ish species (Renau e al., 2010).
Bioin o ma ic Analysis
Fil e ing o he ini ial quali y o he 454 sequences in s o ma
was pe o med using he Roche Newble p og amme. Sequence
analysis was pe o med using he high- h oughpu sequencing
module o CLC Genomics Wo kbench ( e sion 7.5.1; CLC bio,
Aa hus, Denma k). The aw eads we e cleaned by imming
low quali y sequences wi h quali y sco es o <20. Te minal
nucleo ides ( i e nucleo ides a each ex emi y 5′and 3′),
ambiguous nucleo ides, adap e sequences and eads <15 base
pai s (bp) we e disca ded. Fo de no o assembly, con igs <200 bp
we e also disca ded and he de aul local alignmen se ings we e
used o ank po en ial ma ches (misma ch cos o 2, inse ion cos
o 3, dele ion cos o 3). The highes sco ing ma ches ha sha ed
≥50% o hei leng h wi h ≥80% o simila i y we e included in
he alignmen . The assembled ansc ip s we e subjec ed o cd-
hi -es p og amme wi h an iden i y h eshold o 90% o emo e
edundancy (Li and Godzik, 2006; Duan e al., 2012). In o de
o emo e any mi ochond ial and ibosomal con amina ion,
sequences we e compa ed agains pacu mi ochond ial genome
and zeb a ish ibosomal RNA Re Seqs (NCBI da abase) using
CLC Genomic Wo kbench ( e sion 8.0.3; CLC Bio, Aa hus,
Denma k).
Func ional anno a ion o he unique consensus sequences was
pe o med by homology sea ches agains he Na ional Cen e
o Bio echnology In o ma ion (NCBI) non- edundan p o ein
da abase (n ) (cu o E- alue o 1E-3) using BLAST2GO so wa e
(Conesa e al., 2005) o ob ain he pu a i e gene iden i y. All
BLASTx hi s we e il e ed o edundancy in p o ein accessions.
The gene on ology (GO) e ms we e assigned o each unique gene
based on he GO e ms anno a ed o he co esponding homologs
in he NCBI da abase (e- alue cu o 1e-6). The ansc ip s
we e u he anno a ed in In e P o, Enzyme code (EC), and
Kyo o Encyclopedia o Genes and Genomes (KEGG) me abolic
pa hways analysis h ough he Bi-di ec ional Bes Hi s (BBH)
me hod.
Mic osa elli es we e iden i ied in he con igs using
msa commande so wa e (Fai clo h, 2008). P ime s lanking
he mic osa elli e loci we e designed wi h P ime 3plus so wa e
(Rozen and Skale sky, 2000). The six possible eading ames o
he consensus sequence o each unc ionally anno a ed con ig
con aining mic osa elli e we e compa ed agains he NCBI
p o ein da abase using BLASTx (e- alue 1e-10) in o de o
ind Open Reading F ame (ORF) egions. These app oaches
allowed us o loca e mic osa elli es in coding sequences (CDS)
o un ansla ed egions (5′UTR and 3′UTR) h ough g aphical
sequence iewe Table (Milne e al., 2013).
Mic osa elli e Geno yping and Valida ion
DNA was ex ac ed om in agmen s using he Wiza d
Genomic DNA Pu i ica ion Ki (P omega), acco ding o he
manu ac u e ’s p o ocol. Mic osa elli e alida ion was pe o med
in 30 loci, selec ed acco ding o he mo i and unc ional
anno a ion o he con igs. Ampli ica ions we e pe o med by
polyme ase chain eac ion (PCR) in a o al olume o 25 µl
con aining 100 µM o each dNTP (dATP, dTTP, dGTP, and
dCTP), 1.5 mM MgCl2, 1X Taq DNA bu e (20 mM T is-HCl,
pH 8.4, and 50 mM KCl), 0.1 µM o each p ime , 0.5 uni s o
Taq Polyme ase (In i ogen) and 10-50 ng o genomic DNA.
The eac ions we e pe o med in a he mocycle (P oFlexTM
PCR Sys em, Li e Technologies) ollowing ini ial dena u ing o
10 min a 95◦C; 35 cycles o 30 s a 95◦C, 30 s a 55–60◦C
(adjus ed o each p ime se ), 20 s a 72◦C; and a inal ex ension
a 72◦C o 20 min.
Mic osa elli es ha showed polymo phism in 6%
polyac ylamide gels we e analyzed in a 3130xl sequence
(Li e Technologies) o ge be e accu acy o allele de e mina ion.
The sequencing s a egy adop ed in his s udy was acco ding
o p o ocols desc ibed by Schuelke (2000), using he CAG ag
p ime (5′-CAGTCGGGCGTCATCA-3′;Shi k e al., 2013)
labeled wi h he luo och omes HEX o FAM. The geno yping
F on ie s in Gene ics | www. on ie sin.o g 3Feb ua y 2018 | Volume 9 | A icle 46
Jo ge e al. Gene ic Resou ces in Pi api inga
PCR was pe o med wi h he ollowing eagen s: 100 µM o
each dNTP, 1.5 mM MgCl2, 1X Taq DNA bu e , 0.1 µM o each
p ime (F and R), 0.01 µM o he CAG ag p ime , 0.5 uni s o Taq
Polyme ase (In i ogen), and 10–50 ng o genomic DNA. The
cycling p og amme o ampli ica ion consis ed o : nine cycles a
95◦C o 30 s, 55–60◦C o 30 s (adjus ed o each p ime se ),
72◦C o 20 s; hen, 30 cycles a 95◦C o 30 s, 50◦C o 30 s,
and 72◦C o 20 s. Du ing he i s nine cycles, he annealing
empe a u e o 55–60◦C allows inco po a ion o he p ime s
(F and R) om he mic osa elli e loci. Then, in he ollowing
30 cycles, he empe a u e o 50◦C acili a es he annealing
o he luo escen dye-labeled CAG ag p ime . PCR p oduc s
we e analyzed by capilla y elec opho esis wi h a 3130xl gene ic
analyze , using he DS-30 ma ix, wi h he GeneScan 500 ROX
dye Size S anda d (The mo). The p og amme GeneMappe 3.7
(Applied Biosys ems) was used o de e mine he allele sizes.
Mic osa elli e Di e si y and Popula ion
Analysis
Fo s a is ical analysis, we ini ially used GenAlex analysis 6.1
so wa e (Peakall and Smouse, 2012) o con e he a ays
in o speci ic o ma s o each p og amme. The obse ed (Ho)
and expec ed (He) he e ozygosi y, Ha dy-Weinbe g Equilib ium
(HWE) and Analysis o Molecula Va iance (AMOVA) (Exco ie
e al., 1992) we e calcula ed using he A lequim 3.5 p og amme
(Exco ie and Lische , 2010). The le els o signi icance o he
HWE es we e adjus ed wi h he Bon e oni co ec ion (Rice,
1989). The inb eeding coe icien (FIS) was pe o med using
Genepop 4.0.11 (Rousse , 2008), based on Wei and Cocke ham
(1984) es ima es. The ixa ion index (FST) was calcula ed using
FSTAT 9.3.2 so wa e (Goude , 1995). W igh (1965) h eshold
alues we e adop ed, FST =li le gene ic di e en ia ion (0–
0.05); mode a e gene ic di e en ia ion (0.05–0.25); high le el o
gene ic di e en ia ion (>0.25). The p og amme Ce us .3.0.7
(Ma shall e al., 1998) was applied o e i y he p esence o
null alleles. Linkage disequilib ium (LD) was es ima ed using
A lequin .3.5.2.2. The le els o signi icance we e adjus ed o
mul iple es s using he Bon e oni co ec ion.
A e LD analysis, le el o admix u e among popula ion
samples was in e ed by es ima ing he op imum numbe o
clus e s (K), as sugges ed by E anno e al. (2005), using
he p og amme STRUCTURE e sion 2.3.4 (P i cha d e al.,
2000) wi hou p io in o ma ion abou popula ion. P ima ily,
we de e mined he dis ibu ion o 1K, an ad hoc s a is ic
based on he a e o change in he log p obabili y o da a
be ween successi e K alues. The ange o clus e s (K) was
p ede ined om 1 o 5. The analysis was pe o med in 25
eplica ed uns using 200,000 i e a ions a e a bu n-in pe iod o
50,000 uns. The K alue mos likely o explain he popula ion
s uc u e is he modal alue o his 1K. The ou pu s o
STRUCTURE analysis we e isualized h ough he STRUCTURE
HARVESTER p og amme (Ea l, 2012).
Analysis o popula ion bo lenecks was es ed using
BOTTLENECK (Co nue and Luika , 1996; Pi y e al., 1999), by
using he mu a ion–d i equilib ium assuming he wo-phase
model (TPM) wi h 70% s epwise mu a ion model (SMM)
and 30% in ini e allele model (IAM). De ia ions be ween he
obse ed and expec ed equency dis ibu ions we e es ed using
he Wilcoxon’s signed ank es . BOTTLENECK was un o
10,000 i e a ions.
RESULTS
T ansc ip ome Sequencing
The esul s o li e ansc ip ome sequencing in pi api inga
yielded a o al o 192,373 eads, which we e deposi ed in he
Sho Read A chi e (SRA) o NCBI unde he accession numbe
SRR6303971. The aw eads p esen ed an a e age leng h o 395.5
bp, comp ising a o al o ∼76 Mbp. A e he imming p ocess,
he a e age leng h o he eads was o 362.1 bp, esul ing in a
o al o ∼69 Mbp (192,077 eads; Table 1). As P. b achypomus
is conside ed a non-model o ganism, and he e o e wi hou
e e ence genome, de no o assembly s a egy was pe o med
o ansc ip ome analysis, which yielded 3,696 non- edundan
con igs as a esul o 174,272 o e lapping eads (63,460,229 bp).
The size cha ac e is ics o he con igs a e p esen ed in Table 1.
A o al o 17,805 emaining eads (6,084,530 bp) was conside ed
as single ons, and he e o e hey we e no used o subsequen
analysis.
Non- edundan sequences we e anno a ed by BLASTx
algo i hm agains he NCBI da abases: non- edundan p o ein
(n ), p o ein Re Seq o zeb a ish and ugu. A o al o 2,568
unique p o ein accessions (69.4% o ansc ip s) had signi ican
simila i y in he n da abase. In ela ion o he p o ein Re Seq o
zeb a ish and ugu, we ound simila numbe s o anno a ed genes,
which we e o 2,498 (67.6%) and 2,419 (65.4%), espec i ely.
No sequence showed homology wi h known pi api inga p o ein
sequences deposi ed in NCBI da abase, because he a ailable
sequences da abase is s ill limi ed o mos ly mi ochond ial
sequences.
O he 2,568 con igs wi h co espondence in he n da abase,
2,075 (80.8%) we e anno a ed in he ca ego ies o Gene On ology
(GO). A o al o 1,831 assignmen s o Biological P ocess (88.2%)
we e ound, ollowed by 1,757 o Molecula Func ion (84.6%)
and 1,378 o Cellula Componen (66.4%). In ela ion o he
GO subca ego ies, he mos abundan e ms we e ela ed o:
me abolic p ocess, cellula p ocess, and single-o ganism p ocess
o he Biological P ocess ca ego y; binding, ca aly ic ac i i y, and
TABLE 1 | Da a o de no o assembly om li e ansc ip ome o pi api inga
Pia ac us b achypomus.
Ma ched eads o assembly 174,272
To al nucleo ides o ma ched eads 63,460,229
Numbe o con igs 3,797
To al o con ig nucleo ides 2,999,680
Minimum con ig leng h (bp) 202
Maximum con ig leng h (bp) 7,812
A e age con ig leng h (bp) 790
N75 (bp) 593
N50 (bp) 861
N25 (bp) 1,383
F on ie s in Gene ics | www. on ie sin.o g 4Feb ua y 2018 | Volume 9 | A icle 46
Jo ge e al. Gene ic Resou ces in Pi api inga
anspo e ac i i y o he Molecula Func ion; cell, o ganelle,
and memb ane o he Cellula Componen (Figu e 1). In he
p esen s udy, genes assigned o he immune sys em, g ow h
and ep oduc ion we e ound, and he e o e hese da a will
se e as suppo o u u e s udies on he aquacul u e o
pi api inga.
The ansc ip s cha ac e iza ion in he KEGG da abase
demons a ed ha 1,122 sequences we e iden i ied in 106
me abolic pa hways. Genes in ol ed in he biosyn hesis o
an ibio ics, pu ine me abolism and glycolysis/gluconeogenesis
we e able o be highligh ed (Figu e 2).
Mic osa elli e Di e si y and Popula ion
Analysis
The sea ch o sho sequence epea s (SSR) in he 3,696 con igs
esul ed in he disco e y o 130 mic osa elli e ma ke s dis ibu ed
in 95 con igs. In o al, 75 pai s o p ime s we e designed
adjacen o he mic osa elli e loci, including he ollowing
sequence epea s: 56 di, 13 i, 4 e a, and 2 pen anucleo ide.
Among he dinucleo ide mo i s, he main epea s we e he
ypes AC (48.28%), AG (39.65%), AT (10.35%), and CG
(1.72%). In ela ion o he inucleo ide mo i s, we iden i ied
se en ypes (AGC, AGG, ATC, AAT, ACG, CCG, and AAG).
The e anucleo ide (ATCT, AAAG, AATG, and AAAT) and
pen anucleo ide (ACTAT and ATAGT) sequences we e desc ibed
wi h he p esence o ou and wo ypes o mo i s. In ela ion
o he gene posi ion, 26.76% o he mic osa elli e ma ke s we e
ound in he 3′UTR (un ansla ed egion), 19.71% in he 5′UTR,
and 29.58% in he cds (coding sequence).
In he p ocess o mic osa elli e alida ion, 30 ma ke s
we e e alua ed in 22 samples o pi api inga collec ed om
he wild. O hese ma ke s, eigh mic osa elli e loci showed
polymo phism (GenBank accession numbe s MG595996—
MG596003), e ealed by he p esence o di e en agmen
sizes (Table 2). The numbe o alleles was low, which anged
om 2 (loci C25, C64, C410, and C1832) o 5 (C1376) and
mean o 2.750 ±0.366. The expec ed (He) and obse ed (Ho)
he e ozygosi y in he wild popula ion had an a e age o 0.466
±0.061 and 0.355 ±0.076, espec i ely. Mos o he loci
showed posi i e alues o FIS, excep he locus C64. Th ee
mic osa elli e loci (C13, C25, and C1716) showed signi ican
de ia ion om he Ha dy–Weinbe g Equilib ium (HWE) a e
Bon e oni co ec ion (adjus ed p=0.00625).
The esul s o gene ic a iabili y in a med s ocks e ealed ha
No h ish a ms TO1 and TO2 had highe di e si y han he wild
popula ion, demons a ed by numbe o alleles (mean o 4.500 ±
0.423 and 3.375 ±0.596, espec i ely) and a e age alues o He
(0.589 ±0.033 and 0.488 ±0.044, espec i ely) and Ho(0.520 ±
0.060 and 0.447 ±0.040, espec i ely; Table 3). The Sou heas ish
a ms SP1 and SP2 showed he lowes gene ic a iabili y when
compa ed o o he popula ions, wi h lowe allele numbe (mean
o 2.250 ±0.313 and 3.125 ±0.581, espec i ely), and a e age
o He(0.226 ±0.077 and 0.278 ±0.085, espec i ely; p<0.05)
and Ho(0.259 ±0.103 and 0.251 ±0.079, espec i ely; Table 3).
Mos o he mic osa elli e loci we e cha ac e ized wi h posi i e
alues o FIS, excep o SP1 and SP2. The mean alue o
FIS and null alleles was posi i e in mos popula ions, wi h he
excep ion o SP1 (−0.071 ±0.076 and −0.011 ±0.080). The
majo i y o he ma ke s we e in conco dance o HWE, a e
Bon e oni co ec ion, wi h he excep ion o C25 (TO1, SP1, and
SP2), C64 (SP1 and SP2) and C1376 (SP1) (Table 3). Linkage
disequilib ium was ound be ween he mic osa elli es C410 and
C1005 (p<0.00625). Al hough molecula ma ke s on linkage
disequilib ium we e no applied in gene ic a iabili y s udies,
his in o ma ion can be use ul in u u e analysis o gene ic
mapping.
In bo leneck analyses, e idence o ecen educ ions in
popula ion size (bo leneck) using TPM was no ound, excep
o he wild popula ion o Tocan ins Ri e (p=0.027).
In he e alua ion o he le el o admix u e among s ocks by
STRUCTURE, he model-based clus e ing analyses de ec ed K
=3, allowing he iden i ica ion o 3 main clus e s be ween he
popula ions: G oup 1 (SP1 and SP2), G oup 2 (TO1 and TO2),
and G oup 3 (wild) (Figu e 3).
The global FST was 0.379, which showed high gene ic
di e en ia ion among he popula ions (p<0.05). Pai wise
FST de ec ed a highe gene ic di e en ia ion be ween he wild
popula ion and all a med s ocks, pa icula ly when compa ed
o SP1 (FST =0.538, p<0.05) and SP2 (FST =0.537, p<
0.05). Addi ionally, high gene ic s uc u e was ound be ween
he popula ions om No h and Sou heas B azil, as obse ed
be ween TO2 wi h SP1 (FST =0.463, p<0.05) and TO1 wi h
SP2 (FST =0.380, p<0.05; Table 4). Mo eo e , alues o pai wise
FST a e s ock clus e ing de ec ed a highe gene ic di e en ia ion
when compa ing G oup 1/G oup 2 (FST =0.379, p<0.05),
G oup 1/G oup 3 (FST =0.549, p<0.05), and G oup 2/G oup
3 (FST =0.144, p<0.05).
The esul s o AMOVA showed ha he majo i y o
gene ic a ia ion (29.11%, FCT =0.291, p<0.001) occu ed
be ween g oups (acco ding o STRUCTURE clus e ing), while
he a ia ion among indi iduals wi hin popula ions was only
8.21% (FIS =0.126, p<0.001) and among popula ions wi hin
g oups p esen ed 6.06% o gene ic a ia ion (FSC =0.085,
p<0.001).
DISCUSSION
T ansc ip ome Sequencing
Cu en ly, gene ic esou ces o pi api inga P. b achypomus
a e limi ed only o sequences o he mi ochond ial genome
(Chen e al., 2016). Thus, one o he main esul s o
his s udy was he da a gene a ed h ough ansc ip ome
sequencing, because li le knowledge was a ailable abou
he genes o his species. The e iciency o he Roche/454
sequencing sys em in he unc ional genomics analysis o
pi api inga can be obse ed because o he 3,696 ansc ip s
ha we e gene a ed in his s udy. Acco ding o Seeb e al.
(2011), genome educ ion s a egies o NGS sequencing (e.g.,
ansc ip ome sequencing) a e mo e iable when he objec i e
is o p ospec molecula ma ke s and gene ic in o ma ion o
use in aquacul u e, in a low cos and as way. Roche/454
sequencing echnology is one o he main me hods used in
NGS ansc ip ome o non-model ish (Renau e al., 2010;
F on ie s in Gene ics | www. on ie sin.o g 5Feb ua y 2018 | Volume 9 | A icle 46

Jo ge e al. Gene ic Resou ces in Pi api inga
FIGURE 1 | Resul s o unc ional anno a ion and he assignmen o genes in he GO ca ego ies and subca ego ies.
FIGURE 2 | T ansc ip s cha ac e ized in me abolic pa hways da abase o KEGG enzymes (Kyo o Encyclopedia o Genes and Genomes).
Shin e al., 2012; Calduch-Gine e al., 2013; Mu z e al.,
2013).
The esul s o unc ional anno a ion showed ha he
sequences o pi api inga had a high p opo ion o anno a ed
genes when compa ed o he da abase o zeb a ish and
ugu p o eins. The gene anno a ion allowed iden i ica ion
o genomic egions esponsible o on ogene ic de elopmen
p ocesses, biological egula ion, he immune sys em, and
egions in ol ed in p ocesses o g ow h and ep oduc ion.
Consequen ly, he p esen da a can be used as he basis
o u he biological s udies o o he a eas o aquacul u e
o o u u e b eeding p og ammes. In addi ion, h ough
ansc ip ome sequencing, he disco e y o gene-associa ed
mic osa elli es can be conside ed o be he main esul
which can be applied o pi api inga aquacul u e, as al eady
demons a ed in p e ious s udies o ish (Renau e al., 2010;
Helya e al., 2012; Shin e al., 2012). The use o gene-
associa ed ma ke s becomes e en mo e impo an in he
cons uc ion o gene ic maps (Shin e al., 2012) because, by
compa a i e genomics using ish genome e e ences al eady
sequenced, i is possible o p esume he loca ion o each s udied
locus.
F on ie s in Gene ics | www. on ie sin.o g 6Feb ua y 2018 | Volume 9 | A icle 46
Jo ge e al. Gene ic Resou ces in Pi api inga
TABLE 2 | Cha ac e iza ion o he gene ic di e si y o eigh polymo phic mic osa elli es in he wild popula ion o pi api inga (Pia ac us b achypomus).
Locus Sequence
desc ip ion
Gene
posi ion
Mo i s P ime s 5′−3′TA◦C Size ange Na HoHeP(HWE) FIS F(Null)
C13 dihyd oxy i amin d24-like 3′UTR (AGC)6F: TCTCTTCAAGCCTCCTCTGC
R: ATGCTGCAGCTCCTCCTGT
60◦C 143–149 3 0.545 0.669 0.000 0.188 0.075
C25 cy osolic 5-nucleo idase
3a-like iso o m x 1
3′UTR (AT)11 F: CTTTGTCTGCTTTGGGTCGT
R: CTTAGAAGAATGTGCAAATTGAAA
60◦C 117–120 2 0.000 0.169 0.001 1.000 0.887
C64 sodium-coupled neu al
amino acid anspo e
5′UTR (AAAG)7F: CAAAGCAAACTCAAAAAGGAAAA
R: TGGGAACGTTTAGCATCTCA
55◦C 143–151 2 0.545 0.474 0.650 −0.156 −0.082
C410 apolipop o ein e 3′UTR (AG)8F: CGCACAGGTCTAAAGGCACT
R: CTCCCACACAGTGAAAAGCA
60◦C 125−127 2 0.273 0.359 0.271 0.245 0.125
C1005 apolipop o ein e 3′UTR (AG)7F: AGTTGTTGCACCAAATGCAG
R: CTTGTTCCCTCCCACACAGT
60◦C 137−141 3 0.318 0.369 0.225 0.140 0.090
C1376 u ose-biphospha ase
1-like
5′UTR (AC)10 F: GTGTTACATGGCAGGCGTTT
R: CAAGTGAGACCAAATCCAAGG
60◦C 157–175 5 0.500 0.608 0.185 0.180 0.073
C1716 acuola a p syn hase 16
kda p o eolipid subuni
3′UTR (GT)7F: AACCGAAGAGAGGGGAGTGT
R: GCATTTACAAGGGGACGCAC
60◦C 155–171 3 0.545 0.659 0.000 0.175 0.059
C1832 – – (AC)6F: GGTGCTATGTCGTAGAGGCC
R: AGGAAGGCATGACCAGTGTG
60◦C 159–169 2 0.111 0.529 0.036 0.800 ND
The wild popula ion analyzed co esponds o 22 indi iduals collec ed on he Tocan ins Ri e . TA,annealing empe a u e (◦C); Na, numbe o alleles pe locus; Ho,obse ed he e ozygosi y;
He,expec ed he e ozygosi y; P (HWE), p- alue om Ha dy-Weinbe g equilib ium and FIS, inb eeding coe icien . F(Null), Null alleles; ND, no pe o med.
Mo eo e , some examples ha e demons a ed ha
gene-linked mic osa elli e ma ke s can be co ela ed wi h
in e es ing p oduc i e ai s, especially o g ow h pe o mance.
In he ish Spa us au a a, a dinucleo ide mic osa elli e in he
5′UTR o he g ow h ho mone gene (GH) is linked wi h
as e g ow h a e, especially he alleles 250 and 254, which
can be used o b eeding managemen and gene ic selec ion
o his ai (Almuly e al., 2005). In o he ish species, such as
O eoch omis nilo icus and La es calca i e (Yue e al., 2001; Yue
and O ban, 2002), mic osa elli es ha e also been epo ed o
genes o in e es (p olac in, GH and ig 2) and, he e o e, hey
can be used in ma ke -assis ed selec ion (MAS) p og ammes.
In he p esen s udy, eigh polymo phic mic osa elli e loci we e
alida ed, some o hem loca ed in gene egions ha may be
use ul o p oduc i e cha ac e is ics in aquacul u e. In his case, a
mic osa elli e locus was ound in he gene Te aspanin−3 iso o m
x1 (C1832), which plays a ole in i al in ec ion pa hology
(Ma in e al., 2005; Shoshana and Shoham, 2005). The e is
ano he mic osa elli e in he gene Cy osolic 5 – nucleo idase 3
a-like (NTC5C3) (C25), which con ibu es in he p oduc ion
o ed blood cells and i s mu a ion can cause hemoly ic anemia
and in luence on he immune sys em (Aksoy e al., 2009).
Thus, he mic osa elli es desc ibed in his s udy will be also
impo an in u u e analysis o (QTL) linked o ai s o disease
esis ance, which has ecei ed special a en ion in aquacul u e
species, such as u bo (Scoph halmus maximus), ainbow ou
(Onco hynchus mykiss), salmon (Salmo sala ), Nile ilapia (O.
nilo icus), and cod (Gadus mo hua), in es iga ing he esis ance
o pa hogens (Pa do e al., 2008; Ødegå d e al., 2010, 2011;
Yáñez e al., 2014; E enhuis e al., 2015). Fu he mo e, one
mic osa elli e locus was also de ec ed in he gene apolipop o ein
e(C410), which is associa ed wi h he cen al ne ous sys em
and he senescence p ocess (Wang e al., 2014). These ma ke s
can p o ide use ul in o ma ion o s udies o he biology o
he pi api inga, besides se ing as a amewo k o o he na i e
species.
Popula ion Analysis
The alida ion o eigh mic osa elli es showed a low le el
o gene ic di e si y in hese loci, bo h in wild and a med
s ocks. In he wild, he obse ed he e ozygosi y (Ho) anged
om 0.000 o 0.545 and an a e age o 2.750 alleles pe
locus. These alues con i m he low gene ic a iabili y when
compa ed wi h ela ed species, such as pacu P. mesopo amicus
(Ho ange om 0.068 o 0.911 and a e age o 8.5 alleles pe
locus), and ambaqui C. mac opomum (Ho ange om 0.430
o 0.880 and a e age o 12.8 alleles pe locus; Calcagno o
and DeSalle, 2009; Fazzi-Gomes e al., 2017). In con as
o neu al ma ke s (mic osa elli es in noncoding egions),
gene-associa ed mic osa elli es migh be mo e suscep ible o
selec ion p essu e and, he e o e, hey ha e low alues o gene
di e si y.
Analysis o he gene ic di e si y in pi api inga a med s ocks
showed signi ican di e ences be ween ish a ms in di e en
egions o B azil, wo om he Sou heas (São Paulo S a e:
SP1 and SP2) and wo om he No h (Tocan ins S a e: TO1
and TO2). In gene al, a med s ocks we e expec ed o ha e
low gene ic a iabili y as a esul o gene ic decline, gene ic
F on ie s in Gene ics | www. on ie sin.o g 7Feb ua y 2018 | Volume 9 | A icle 46
Jo ge e al. Gene ic Resou ces in Pi api inga
TABLE 3 | Values o gene ic di e si y o eigh mic osa elli e loci o Pia ac us b achypomus.
S ocks Loci
C13 C25 C64 C410 C1005 C1376 C1716 C1832
Wild N 22 22 22 22 22 22 22 9
Na 3 2 2 2 3 5 3 2
Ho0.545 0.000 0.545 0.273 0.318 0.500 0.545 0.111
He0.669 0.169 0.474 0.359 0.369 0.608 0.659 0.529
P(HWE) 0.000 0.001 0.650 0.271 0.225 0.185 0.000 0.036
FIS 0.188 1.000 −0.156 0.245 0.140 0.180 0.175 0.800
F(Null) 0.075 0.887 −0.082 0.125 0.090 0.073 0.059 ND
TO1 N 25 25 25 25 25 25 25 25
Na 4 5 3 3 4 6 6 5
Ho0.600 0.160 0.560 0.520 0.560 0.440 0.760 0.560
He0.536 0.704 0.495 0.537 0.562 0.609 0.782 0.580
P(HWE) 0.650 0.001 0.365 0.267 0.421 0.010 0.545 0.388
FIS −0.121 0.776 −0.133 0.034 0.004 0.282 0.028 0.035
F(Null) −0.067 0.620 −0.065 −0.015 −0.034 0.164 0.011 −0.001
TO2 N 26 21 26 26 26 26 26 26
Na 3 2 2 2 2 6 5 5
Ho0.500 0.380 0.346 0.384 0.384 0.538 0.461 0.653
He0.528 0.315 0.382 0.506 0.506 0.632 0.515 0.638
P(HWE) 0.840 1.000 0.626 0.256 0.256 0.020 0.030 0.670
FIS 0.055 −0.212 0.096 0.244 0.244 0.151 0.107 −0.253
F(Null) 0.018 −0.103 0.040 0.127 0.127 0.055 0.067 −0.027
SP1 N 14 14 14 14 14 14 14 14
Na 2 1 1 3 2 3 3 3
Ho0.143 – – 0.287 0.500 0.857 0.143 0.143
He0.137 – – 0.264 0.494 0.634 0.140 0.203
P(HWE) 1.000 – – 1.000 1.000 0.000 1.000 0.109
FIS −0.040 – – −0.083 −0.011 −0.368 −0.019 0.306
F(Null) −0.028 – – −0.069 −0.023 −0.215 −0.027 0.272
SP2 N 19 19 19 19 19 16 18 19
Na 2 2 1 3 3 5 3 6
Ho0.368 0.000 – 0.157 0.157 0.625 0.222 0.473
He0.308 0.193 – 0.152 0.152 0.790 0.207 0.486
P(HWE) 1.000 0.002 – 1.000 1.000 0.112 1.000 0.387
FIS −0.200 1.000 – −0.038 −0.385 0.214 −0.070 0.027
F(Null) −0.099 0.916 – −0.032 −0.032 0.093 −0.051 0.024
Wild, popula ion om he Tocan ins Ri e ; TO1 and TO2, ish a ms om Tocan ins; SP1 and SP2, ish a ms om São Paulo, Na, numbe o alleles; Ho, obse ed he e ozygosi y; He,
expec ed he e ozygosi y; P (HWE), P- alue om Ha dy-Weinbe g equilib ium; FIS,inb eeding coe icien ; F(Null), Null alleles; ND, no pe o med.
d i , selec ion and inb eeding (Theodo ou and Cou e , 2015).
Howe e , he esul s o his s udy showed highe gene ic
a iabili y in b eeding s ocks om No h ish a ms in ela ion
o he wild s ocks (p<0.05; highe alues o allelic equency
and he e ozygosi y), which was also obse ed in s udies wi h
o he ela ed species (Ba oso e al., 2005; Pana a i-An unes e al.,
2011). The basis o his esul could be conside ed om h ee
di e en pe spec i es: (1) No h ish a ms had o igina ed om
di e en wild s ocks esul ing in high le el o gene ic a iabili y;
(2) p oblems o sample size bias, such as ew mic osa elli e
loci and indi iduals analyzed; (3) e idence o ecen gene ic
bo lenecks in he wild popula ion. Some s udies o ish ha e
epo ed bo lenecks in na u al popula ions, pa icula ly due o
habi a loss and agmen a ion by human dis u bance (B aue
e al., 2016). In he case o pi api inga, he agmen a ion o
he Tocan ins Ri e by hyd oelec ic dams in he 80′s and
90′s (e.g., Tucu uí and Luiz Edua do Magalhães dams, whe e
wild ish we e collec ed o his s udy) could be esponsible o
F on ie s in Gene ics | www. on ie sin.o g 8Feb ua y 2018 | Volume 9 | A icle 46
Jo ge e al. Gene ic Resou ces in Pi api inga
FIGURE 3 | E alua ion o he le el o admix u e among s ocks by
STRUCTURE, showing h ee main clus e s be ween he popula ions: G oup 1
(SP1 and SP2) in g een, G oup 2 (TO1 and TO2) in ed, and G oup 3 (wild) in
yellow.
TABLE 4 | Analysis o pai wise FST based on eigh mic osa elli e loci be ween
popula ions o Pia ac us b achypomus.
SP1 SP2 TO1 TO2 wild
SP1 –
SP2 0.18438 –
TO1 0.39201 0.38080 –
TO2 0.46373 0.42999 0.08345 –
wild 0.53856 0.53782 0.16086 0.17915 –
Wild, popula ion om he Tocan ins Ri e ; TO1 and TO2, ish a ms om Tocan ins; SP1
and SP2, ish a ms om São Paulo. All esul s o FST we e signi ican s a is ically p <0.05.
a popula ion educ ion and subsequen gene ic a ia ion loss
de ec ed by ou mic osa elli e analysis. The e a e conside able
numbe s o hyd opowe dams in he basin, which can a ec
he ep oduc ion, mig a o y ou es, and egg and la ae d i
o ish (Agos inho e al., 2008). Al e a ion o he mig a o y
low consequen ly leads o a dec ease in o in e up ion o
he gene low, educing he popula ion size, which makes he
ish mo e suscep ible o he e ec s o gene ic d i (Ha anaka
and Gale i, 2003), which esul s in gene ic s uc u e o some
ish species (Calcagno o and DeSalle, 2009; Do P ado e al.,
2018).
STRUCTURE and pai wise FST analyses sugges ed a
high gene ic s uc u e be ween he s ocks he ein analyzed,
pa icula ly as esul o he ixa ion o speci ic alleles in some
loci, which esul ed in h ee clus e s (Figu e 3). The e a e
h ee hypo he ical explana ions o hese gene ic pa e ns:
(1) di e en ia ion o wild popula ion in ela ion o a med
s ocks, which could be due o he selec ion o he i es
indi iduals o a ming sys ems o low numbe o ounde s
o he es ablishmen o he a med b oods ocks; (2) lowe
gene ic s uc u e in No h/wild han Sou heas /wild, which
sugges s ha No h ish a ms had equen b oods ock
eno a ion om he wild; (3) ish a ms we e gene ically
clus e ed due o he geog aphic dis ibu ion, i.e., he deg ee
o gene ic simila i y is highe when one ish a m is close
o he o he , indica ing in e change o indi iduals be ween
nea by ish a ms, common o igin o he a med b oods ocks,
o ixa ion/selec ion o speci ic alleles o di e en clima ic
condi ions ha a e ound in B azil (No h and Sou h). Howe e ,
hese gene ic pa e ns should be also e alua ed using neu al
ma ke s (mic osa elli es in noncoding egions) and h ough
echniques o highe genome co e age (SNP, single-nucleo ide
polymo phism).
Th ough AMOVA analysis, he main gene ic a ia ion was
ound o be p esen wi hin popula ions (64.8%). This gene ic
pa e n has also been epo ed in s udies ca ied ou wi h
pacu (Calcagno o and DeSalle, 2009; Ie olino e al., 2010)
and ambaqui (Aguia e al., 2013). Mo eo e , highly signi ican
gene ic a ia ion was associa ed wi h di e ences be ween g oups
(Wild, SP, and TO), which ep esen ed 29.11% o gene ic
a ia ion, in con as o low di e ences among popula ions
wi hin g oups (6.06%).
In gene al, ou s udy o gene ic cha ac e iza ion in pi a ininga
a med s ocks p o ides impo an insigh s which can lead
o be e managemen o his species in aquacul u e. Ou
esul s a e undamen al o beginning a b eeding p og amme,
since he gene ic s uc u e should be aken in o conside a ion
when composing an ini ial base popula ion, whe e ma ings
be ween a med indi iduals om No h and Sou heas B azil
a e shown o esul in highe gene ic a iabili y in he amilies.
Mo eo e , he da a sugges ed le els o gene ic di e si y which
we e highe in a med s ocks han in wild ish, disca ding he
occu ence o inb eeding. In gene al, lack o knowledge on
gene ic a iabili y o s ocks can esul in inb eeding and ixa ion
o dele e ious genes, educed g ow h a es, disease esis ance
p oblems and educed abili y o adap o new en i onmen s
(A kush e al., 2002; Galla do e al., 2004; Nei a e al.,
2006; Hillen e al., 2017). The e o e, besides he iden i ica ion
o QTL o assis in he selec ion o supe io geno ypes by
MAS, s udies o mic osa elli es a e impo an o gene ic
moni o ing, suppo ing pi api inga aquacul u e and inc easing
i s p oduc i i y.
FINAL CONSIDERATIONS
The p ospec ion o gene ic da a o pi api inga is one o
he p io i y issues o aquacul u e, since his species is
o high economic impo ance in na ional and global ish
a ming. The iden i ica ion o gene-associa ed mic osa elli es
by NGS is undamen al o unde s anding he gene ic
s uc u e o wild and a med popula ions, p o iding
suppo o u he managemen p og ammes and gene ic
p e-b eeding p og ammes. Mo eo e , he mic osa elli es
desc ibed he ein a e in e es ing a ge s used o ind QTL
ma ke s, speci ically ela ed o he immune sys em o
pi api inga.
AUTHOR CONTRIBUTIONS
PJ: Acquisi ion, analysis and in e p e a ion o da a, d a o
he wo k, inal app o al o he e sion; VM-F, RA, and MdF:
D a o he wo k, de elopmen o in ellec ual con en , inal
app o al o he e sion; MH: Analysis and in e p e a ion o da a,
d a o he wo k, inal app o al o he e sion; NM: Analysis
and in e p e a ion o da a, d a o he wo k, inal app o al
o he e sion; MV: Analysis and in e p e a ion o da a, d a
F on ie s in Gene ics | www. on ie sin.o g 9Feb ua y 2018 | Volume 9 | A icle 46