FunMappOne : a tool to hierarchically organize and visually navigate functional gene annotations in multiple experiments

Scala e al. BMC Bioin o ma ics (2019) 20:79 h ps://doi.o g/10.1186/s12859-019-2639-2 SOFTWARE Open Access FunMappOne: a ool o hie a chically o ganize and isually na iga e unc ional gene anno a ions in mul iple expe imen s Gio anni Scala1,2,3, Angela Se a1,2, Vee Singh Ma wah1,2, Lau a Aliisa Saa imäki1,2 and Da io G eco1,2,3 * Abs ac Backg ound: Func ional anno a ion o genes is an essen ial s ep in omics da a analysis. Mul iple da abases and me hods a e cu en ly a ailable o summa ize he unc ions o se s o genes in o highe le el ep esen a ions, such as on ologies and molecula pa hways. Anno a ing esul s om omics expe imen s in o unc ional ca ego ies is essen ial no only o unde s and he unde lying egula o y dynamics bu also o compa e mul iple expe imen al condi ions a a highe le el o abs ac ion. Se e al ools a e al eady a ailable o he communi y o ep esen and compa e unc ional p o iles o omics expe imen s. Howe e , when he numbe o expe imen s and/o en iched unc ional e ms is high, i becomes di icul o in e p e he esul s e en when g aphically ep esen ed. The e o e, he e is cu en ly a need o in e ac i e and use - iendly ools o g aphically na iga e and u he summa ize anno a ions in o de o acili a e esul s in e p e a ion also when he dimensionali y is high. Resul s: We de eloped an app oach ha exploi s he in insic hie a chical s uc u e o se e al unc ional anno a ions o summa ize he esul s ob ained h ough en ichmen analyses o highe le els o in e p e a ion and o map gene ela ed in o ma ion a each summa ized le el. We buil a use - iendly g aphical in e ace ha allows o isualize he unc ional anno a ions o one o mul iple expe imen s a once. The ool is implemen ed as a R-Shiny applica ion called FunMappOne and is a ailable a h ps://gi hub.com/g ecolab/FunMappOne. Conclusion: FunMappOne is a R-shiny g aphical ool ha akes in inpu mul iple lis s o human o mouse genes, op ionally along wi h hei ela ed modi ica ion magni udes, compu es he en iched anno a ions om Gene On ology, Kyo o Encyclopedia o Genes and Genomes, o Reac ome da abases, and epo s in e ac i e maps o unc ional e ms and pa hways o ganized in a ional g oups. FunMappOne allows a as and con enien compa ison o mul iple expe imen s and an easy way o in e p e esul s. Keywo ds: Func ional anno a ion, Pa hway isualiza ion, On ology isualiza ion, KEGG, Gene On ology, Reac ome, R-Shiny Backg ound Func ional anno a ion o la ge se s o signi ican genes is o en he inal s ep o omics da a analysis. Howe e , when mul iple genes a e selec ed du ing di e en ial anal- ysis, i becomes almos impossible o unde s and he al e ed biological p ocesses by manually inspec ing he *Co espondence: [email p o ec ed] 1Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, A o Ylpön ka u 34 - A o building, FI-33014 Tampe e, Finland 2BioMediTech Ins i u e, Uni e si y o Tampe e, A o Ylpön ka u 34 - A o building, FI-33014 Tampe e, Finland Full lis o au ho in o ma ion is a ailable a he end o he a icle indi idual genes. This ask is e en mo e di icul when compa ing unc ional p o iles de i ed om wo o mo e ela ed expe imen s a he gene le el, o di e en se s o unc ionally ela ed genes may be speci ically a ec ed in di e en expe imen al condi ions. A mul i ude o ools a e al eady a ailable o he com- muni y o g aphically ep esen en iched unc ional anno- a ions om single pai -wise compa isons [1–4]. When conside ing mul iple expe imen s, hese me hods equi e o un sepa a e analyses o each expe imen and subse- quen ly colla e he esul s o compa ison. The complexi y o his ask inc eases wi h he numbe o conside ed © The Au ho (s). 2019 Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Scala e al. BMC Bioin o ma ics (2019) 20:79 Page 2 o 7 expe imen s, especially o use s who a e no amilia wi h ad anced echniques o da a manipula ion. Some ools allow he isualiza ion o he en iched Gene On ol- ogy e ms om mul iple expe imen s [5–7]. Howe e , as hey a e ypically implemen ed in R, hey equi e a ce - ain deg ee o p og amming expe ise in o de o p oduce he desi ed isualiza ions. Mo eo e , since hese me hods usually o e a s a ic g aphical ou pu , he p oduced plo s become di icul o ead and in e p e when la ge numbe o unc ional e ms need o be displayed. An impo an aspec o some unc ional anno a ions is he possibili y o de i e a hie a chical s uc u e o hei base e ms, such as o Kyo o Encyclopedia o Genes and Genomes (KEGG) pa hways [8], Reac ome pa hways [9] and Gene On ology e ms [10]. This s uc u e can be used o o ganize he unc ional e ms and summa ize se s o ela ed unc ions in supe classes. This ea u e can be u he exploi ed o educe he dimensionali y o se s o en iched e ms and o abs ac he unde lying biological unc ions o highe le els o in e p e a ion. He e we p esen FunMappOne, an R-shiny use - iendly so wa e wi h a simple g aphical in e ace ha akes in inpu lis s o human o mouse genes om mul iple expe i- men s, op ionally wi h hei gene-associa ed me ics, such as old change and p- alue. I p o ides unc ionali ies o s a is ically e alua e o e - ep esen ed biological e ms om Gene On ology, KEGG, o Reac ome da abases, g aphically summa ize, and na iga e hem. Me hod The h ee-le el hie a chy In o de o educe he dimensionali y o he se s o en iched e ms, we in oduced he concep o hie a chical summa iza ion, ha is he possibili y o explo e en iched e ms a highe unc ional le els. To do his, a hie a chy is needed o g oup e ms in supe -classes. By de ini ion, his s uc u e needs o be ep esen ed as a di ec acyclic g aph, wi h a oo ca ego y ( ep esen ing he unc ional anno a- ion) and a se ies o me a- e ms ( eal e ms o unc ional g oups), de ining p og essi ely specialized g oup o e ms. This s uc u e is na u ally ound in he in insic o gani- za ion o KEGG and Reac ome pa hways while i can be easily de i ed o Gene On ology e ms, as desc ibed in he nex sec ion. An impo an ac o o he hie a chy de ini ion and cons uc ion is he numbe o le els o he hie a chy, namely he dep h o he co esponding g aph s uc u e: KEGG has an in insic s uc u e based on h ee le els, while Reac ome pa hways and gene on ology can ha e mo e han h ee le els ha a e no uni o mly dis- ibu ed ( he hie a chical chain o me a- e ms can ha e di e en leng h o di e en e ms). Ha ing many summa- iza ion le els has he ad an age o making mo e special- ized g ouping o e ms bu would also complica e he ask o he use o educe he se dimensionali y and ob ain easie iews o he en ichmen da a. Fo his eason, we chose o ollow he KEGG philosophy and homogenize he h ee hie a chies (KEGG, Reac ome and Gene On ology) in o de o ha e h ee le els o summa iza ion om he e ms o he oo . The de ailed implemen a ion o he hie a chies is desc ibed in he ollowing sec ion. Hie a chy de ini ion Figu e 1shows he implemen ed p ocedu e o de ine hie - a chical s uc u es o KEGG pa hways (panel A), Gene On ology e ms (panel B) and Reac ome pa hways (panel C), espec i ely. Fo each anno a ion ype, a h ee-le el hie a chy was de ined. •Fo KEGG pa hways (Fig. 1a), he h ee le els o BRITE unc ional hie a chy was used [8]. •Fo each Gene On ology ca ego y CAT (Biological P ocesses - BP, Cellula Componen s - CC, and Molecula Func ions - MF), a h ee-le el hie a chy was ex ac ed by i s conside ing he g aph GO_CAT oo ed in CAT (Fig. 1b). Then, he acyclic di ec ed subg aph GO_CAT_ac was compu ed by conside ing only he edges ep esen ing he ela ionship “is_a” o “pa _o ” in GO_CAT. Finally, a new g aph GO_CAT_hie was buil by conside ing all he nodes in GO_CAT_ac, and adding, o each node i, all he edges in he pa h [ i,..., −1]i he pa h [ i,..., −1,CAT] o leng h a mos 3 al eady exis ed in GO_CAT_ac. Fo he pa hs [ i,..., −2, −1,CAT] in GO_CAT_ac o leng h g ea e han 3, only he a cs o ming he sequence [ i, −2, −1] we e added o GO_CAT_hie . •Fo he Reac ome pa hways (Fig. 1C), he se o oo nodes Rs we e conside ed and a h ee-le el hie a chy was explica ed. Fi s , he associa ed g aph REACT_RSi oo ed in CAT was selec ed. Nex , o each node i he edges [ i, −1,RSi]we e added i he pa h [ i, −1,RSi]belonged o REACT_RSi.I he pa h [ i,..., −2, −1,RSi]exis ed in REACT_RSi, only he edges o ming he sequence [ i, −2,RSi] we e added o he new g aph ep esen ing he hie a chy. FunMappOne algo i hm wo k low Figu e 2shows he FunMappOne algo i hm wo k low. The inpu is p o ided as N lis s o genes, one o each expe imen al condi ion o compa e and, op ionally, N lis s o modi ica ions (e.g. he old-change o he p- alue) asso- cia ed wi h each gene. Fo each expe imen analyzed, he en iched e ms in he chosen unc ional anno a ion a e compu ed by using he gP o ile R package [4], and a ma ix Te [NxM] is c ea ed, whe e M is he o al numbe o en iched e ms. Each elemen Te [i,j] is associa ed wi h he hype geome ic es p- alue o e m j o he genes Scala e al. BMC Bioin o ma ics (2019) 20:79 Page 3 o 7 ab c Fig. 1 De ini ion o he hie a chies. Fo each unc ional anno a ion ype, a model e lec ing he ela ionship be ween unc ional e ms and le els in hei o iginal s uc u e is shown abo e he co esponding gene a ed hie a chy. Panel a,band c epo hie a chy gene a ion models o KEGG, Gene On ology and Reac ome, espec i ely. In he second and hi d panel, di e en numbe s indica e di e en unc ional e ms. In panel b,“CAT”canbe one o he Gene On ology ca ego ies BP, CC o MF in he i- h lis . Op ionally, Te [i,j] can also be associa ed wi h a alue ha summa izes he modi ica ion alues (e.g. he median old change) o he genes om he i- h lis in e sec ing he gene se o he e m j. To summa ize he in o ma ion a a highe le el o in e - p e a ion, a new ma ix Te i[NxK] is c ea ed, whe e i=1,2 is he desi ed heigh o he chosen anno a ion hie a chy and K is he numbe o di e en e ms a le el i. Each ele- men Te i[i,j] is hen associa ed wi h a summa y s a ics (e.g. he median p- alue) o he elemen s Te [i,k] o all k such ha he e m k is a descendan j in he e e ence hie a chy. Finally, gi en a ma ix Te i[NxK] ep esen ing he en ichmen a le el ias de ined abo e, he possibili y o eo de and clus e expe imen s, based on a gi en dis- ance unc ion Dk,l,isimplemen ed.Thisiscompu ed be ween he ec o s Te i[k,] and Te i[l,] using, al e na- i ely, a dis ance based on he Jacca d index on he num- be o common en iched e ms, he Euclidean dis ance on he alues associa ed wi h e ms, o a combina ion o hese wo. In he i s case, he Jacca d index Jk,lis compu ed as |Te ms(k)∩Te ms(l)| |Te ms(k)∪Te ms(l)|,whe eTe ms(x)is he se o en iched e ms o he expe imen xand Dk,lis se as 1 −Jk,l. In he second case, he se comm(k,l)=Te ms(k)∩ Te ms(l)|is i s conside ed, whe e Te ms(x)is he se o en iched e ms o he expe imen al condi ion x, heni |comm(k,l)|≥0 he Euclidean dis ance DEk,lon he sub- ec o s Te i[k,comm]andTe i[l,comm]iscompu ed.A combina ion o he wo me hods is implemen ed by c e- a ing he mean dis ance ma ix Mk=(D+DE01)/2, whe e Dis hema ixo heJacca dindexandDE01 is he Euclidean dis ance ma ix scaled in he ange [0,1]. In his way, he expe imen al condi ions a e clus e ed oge he no only when hey sha e he same en iched e ms, bu also conside ing how simila a e he en iched e ms wi h espec o hei en ichmen p- alue o summa y s a is ic. A hie a chical clus e ing unc ion is hen applied o he ma ix using a linkage me hod be ween comple e, single and wa d. Resul s and discussion The analy ical app oach p esen ed abo e was imple- men ed using R-shiny. The ypical analysis is pe o med by h ee in e ac ion s eps: i) inpu o gene lis s and modi- ica ions, ii) g aphical isualiza ion o en iched e ms and iii) in e ac i e na iga ion o he esul s. A s ep-by-s ep use manual is a ailable in Addi ional ile 1. In he i s s ep, he applica ion p o ides a simple g aph- ical in e ace, whe e he use can submi a sp eadshee ile wi h he lis s o genes associa ed o each expe imen al condi ion o in e es and (op ionally) hei modi ica ion Scala e al. BMC Bioin o ma ics (2019) 20:79 Page 4 o 7 Fig. 2 FunMappOne wo k low. The ool accep s as inpu gene lis s and modi ica ion alues o e e y expe imen al condi ion S1,...,Sn o which he en ichmen will be ca ied ou . The analysis pe o med on he j- h sample will esul s in a se o en iched e ms Tsj1,...,Tsjk wi h an associa ed p- alue (En .P) om he en ichmen unc ion applied on he gene lis , o a alue coming om he applica ion o a summa y s a is ic (SS) on he associa ed modi ica ion alues. A ma ix wi h n ows associa ed o samples and mcolumns associa ed wi h he en iched e ms is hen speci ied o ep esen he da a s uc u e benea h Le el 3 ep esen a ion o he da a. Ma ices associa ed o highe hie a chical le els a e composed by n ows and as many columns as he ca ego ies o he le el. Each cell o a highe le el ma ix con ains a alue ob ained by applying SS o he e ms belonging o he associa ed ca ego y om he Le el 3 ma ix in o ma ion (e.g. he associa ed old change om a di - e en ial exp ession analysis). The inpu sp eadshee con- ains a shee o e e y expe imen al condi ion, named wi h a condi ion id. In e e y shee , wo columns a e p o- ided, con aining he gene iden i ie s (En ez Gene, Gene Symbol, o Ensembl gene ids) and, op ionally, hei mod- i ica ions, espec i ely. Fu he mo e, an addi ional shee is equi ed, con aining wo columns wi h he condi ion id and he condi ion g ouping in o ma ion, espec i ely. Theuse is henasked ochoose hespecies(human o mouse), a unc ional anno a ion (Gene On ology - BP, Gene On ology - CC, Gene On ology - MF, KEGG, Reac- ome), a summa iza ion unc ion (min, median, mean, max) o anno a e and summa ize he en iched e ms wi h p o ided modi ica ions, a p- alue co ec ion me hod (gSCS [4], bon e oni, d ), and a s a is ical signi icance h eshold o he en iched unc ional e ms. I he ampli- ude o gene modi ica ion (e.g. old change, p- alue) is p o ided, he use selec s whe he he summa ized alue o he en iched e ms is plo ed in a colo -scale associa ed o i s alue, o wi h h ee colo s only (nega i e, ze o, posi- i e); his la e ea u e is use ul when emphasis is gi en o he dominan sign o he modi ica ion in he e m. Mo e- o e , i gene modi ica ion alues a e p o ided in he inpu , heuse canchoose he ypeo in o ma ion ha willbe associa ed o he en iched e ms: he e m en ichmen p- alue, he p o ided modi ica ion alue, o a combina- ion o e m en ichmen p- alues (En .P) and modi ica ion alues (MVs), speci ied as MV ×−log(En .P). Al e na- i ely, i only gene lis s wi hou p o iding modi ica ion alues a e uploaded, he en ichmen p- alue o each en iched e m will be displayed. A e loading he needed iles, a dedica ed panel in he so wa e g aphical en i onmen shows he con en o he p o ided ables, along wi h a summa y o each column. A e clicking he “Gene a e Map” bu on, he ool com- pu es he en ichmen and shows he “Plo Maps” panel. A e selec ing he desi ed isualiza ion op ions and click- ing he “Plo Map” bu on, he ool shows he map o en iched e ms as a g id (Fig. 3), whe e columns ep esen expe imen al condi ions, e en ually g ouped based on he p o ided in o ma ion, and ows ep esen he en iched e ms g ouped and colo ed based on he co esponding hie a chy class. The use can in e ac wi h he gene a ed en ichmen map in h ee di e en ways: i) by selec ing he le el a which he map is displayed, ii) by speci ying one o mo e ca ego ies o e ms o be displayed om a desi ed le el Scala e al. BMC Bioin o ma ics (2019) 20:79 Page 5 o 7 Fig. 3 In e ac i e Map Visualiza ion. The use can selec he le el o hie a chy o isualize (1) as well as a subse o elemen s o be plo ed a each le el o hie a chy (2-4). Fu he mo e, he use can selec a subse o he condi ions (5). In he “Plo sec ion” he use can selec o show he ca ego ies (6) and o keep he aspec a io (7) o he plo . By clicking he bu on “Plo Map” (8) he upda ed map is isualized. A e speci ying he desi ed heigh (9) and wid h (10) o he pd ha will be downloaded, he use can sa e he image by clicking he “Download” bu on (11). Expe imen s can be clus e ed by selec ing he numbe o clus e s (12), he desi ed clus e ing unc ion (13), he dis ance unc ion (14), and hen clicking he “Clus e samples” bu on (15). The map can be ese o he ini ial isualiza ion wi h he p ede ined g ouping by clicking he “Rese clus e ” bu on (16) o hie a chy, iii) by choosing a subse o expe imen al condi ions o be plo ed. The selec ion o he summa- iza ion le el is pe o med ia a d op-down menu. Once he desi ed le el is selec ed and he “Plo Map” bu - on is clicked, he panel wi h he esul s is au oma ically upda ed, p o iding a new map whe e he ows co espond o he ca ego ies o he chosen le el, g ouped by hei supe classes in he hie a chy. The colo o he cells in he new map is associa ed wi h he summa ized alue o all he en iched e ms in he expe imen al condi ion column belonging o he ca ego y ow. The concep o le el ca ego ies can be used o selec subse s o ows o in e es . This is done by selec ing, o each ep esen ed le el, he ca ego ies/ e ms o in e es . The ool subsequen ly upda es he map epo ing only ca - ego y/ e ms om he selec ed se , hus allowing a mo e compac iew o he po ion o in e es o he map. Sim- ila ly, he use can speci y a subse o expe imen s o be plo ed. Finally, he columns o he map can be eo de ed by g ouping expe imen al condi ions ha ing simila en ich- men p o iles. This is accomplished by selec ing a desi ed numbe o g oups, a dis ance unc ion among Jac- ca d, Euclidean and “Jacca d+Euclidean”, and a clus e ing linkage me hod be ween comple e, single, and wa d. In he “Clus e ing” sub- ab o “Plo Maps”, FunMappOne p o ides a isualiza ion o he clus e dend og am as well as he pa i ioning based on he numbe o desi ed clus- e s. This unc ionali y can help in selec ing he mos app op ia e numbe o clus e s o be displayed. Finally, he cu en iew o he map can be expo ed in a ious g aphical o ma s. We inally p o ide a compa ison among FunMappOne ea u es and hose o e ed by a selec ion o cu en ly a ail- able ools o unc ional anno a ion ha ing close scope o FunMappOne. Table 1shows he compa ison o Fun- MappOne wi h he ollowing gene unc ional analysis ools: DAVID [1], En ich [2], ToppGene [3], g:p o ile [4], clus e P o ile [5], Goplo [6]andBACA[7]. As shown in Table 1, mos o he o he ools o e he pos- sibili y o analyze KEGG pa hways, Reac ome pa hways and Gene On ology, also wi h a g aphic ep esen a ion o he en ichmen esul s. Only Goplo o e s he pos- sibili y o map gene associa ed alues o e ms, while En ich and g:p o ile a e he only ools o e ing a web based g aphical use in e ace. None o he o he ools o e he possibili y o summa ize esul s and o clus- e unc ional p o iles om mul iple expe imen s. To Scala e al. BMC Bioin o ma ics (2019) 20:79 Page 6 o 7 Table 1 Compa ison wi h exis ing ools Fea u e/Tool DAVID En ich ToppGene g:p o ile clus e P o ile Goplo BACA FunMappOne KEGG pa hways      Reac ome pa hways      Gene On ology       G aphic ep esen a ion    G aphic use in e ace    Hie achycal summa iza ion  Mul iple expe imen s  Te m based clus e ing  Mapping alues on e ms  Di e en ools a e epo ed on columns, desi ed ea u es a e epo ed on ows. Check-ma ks ep esen he p esence o he ea u e in he ool ou knowledge, FunMappOne is he only ool p o iding all o hese unc ionali ies in a use iendly g aphical in e ace. Case s udy We showcase he unc ionali ies o FunMappOne on a ansc ip ome da ase o mouse hepa ocy es exposed o 26 chemical compounds wi h di e en ca cinogenic po en ial [11]. While Schaap e al. de ined he simila i y be ween he mechanism o ac ion o a pai o chemicals a he le el o indi idual genes, we es ed he hypo hesis ha signi ican simila i y pa e ns can be obse ed also a he unc ional anno a ion le el. An excel ile (Addi ional ile 2) con aining he o iginally desc ibed lis s o he 30 mos up- egula ed and 30 mos down- egula ed genes in each compound- o-con ol compa ison, along wi h he co esponding -s a is ics, was uploaded o FunMappOne. The anno a ion was pe o med by selec ing he “KEGG” op ion and “gSCS” as mul iple es ing co ec ion me hod wi h “0.05” as signi icance h eshold (Addi ional ile 3). Fo he plo ing, he “median” unc ion was chosen as summa y s a is ics and colo s we e associa ed o he summa ized modi ica ion di ec ion o en iched e ms by selec ing he “sign” op ion (Addi ional ile 3). Chemical exposu es we e inally o de ed based on he “Jacca d” dis ance on he numbe o sha ed e ms, and u he clus- e ed in o 11 g oups using hie a chical clus e ing and “comple e” agg ega ion me hod. Addi ional ile 3shows he KEGG en ichmen map a he le el 1 (Addi ional ile 3A), le el 2 (Addi ional ile 3B), and a he indi idual pa hway le el 3 (Addi ional ile 3C). Ou analysis con i med many simila i ies o iginally desc ibed by Schaap and collabo a o s, such as he one be ween Wye h-14643 (WY) and Clo ib a e (CF), which in ou analysis we e g ouped oge he wi h Tac olimus (FK506) in clus e 11 (Addi ional ile 3C). These chem- icals modula e PPAR signalling pa hway and a y acid me abolism ela ed genes, which we obse ed o be signi ican ly en iched. Mo eo e , we iden i ied a la ge clus e o compounds (clus e 6) cha ac e ized by no signi ican ly en iched pa hway, whose pai wise simila i y o hei mechanism o ac ion we e also desc ibed in he o iginal epo , bu wi h a low signi icance [11]. In e es ingly, en iched al e a ion o pa hways ela ed o s e oid ho mone biosyn hesis and chemical ca cino- genesis was obse ed in a g oup o known ca cinogenic compounds clus e ed oge he (clus e 5). The isualiza- ions p oduced a highe le els o he pa hway hie a chy help he use o immedia ely obse e ha he chemicals in clus e 5 al e he genes in me abolic pa hways and human diseases (Addi ional ile 3A). When he isualiza ion a le el 2 is inspec ed, he no ion ha lipid me abolism and cance pa hways a e en iched also easily eme ges. This unc ionali y o FunMappOne becomes e y e ec- i e when analyzing iche unc ional anno a ions, such as gene on ology, whe e he numbe o en iched e ms can be signi ican ly highe (as shown in Addi ional ile 4). Conclusion We p esen FunMappOne, a web based s andalone appli- ca ion ha enables use s o g aphically inspec , na iga e, and compa e unc ional anno a ions in mul iple expe i- men s a di e en le els o abs ac ion. This ool acili a es he analyses o mul iple expe imen al condi ions h ough a simple use in e ace and dynamic g aphical ep esen a- ions o he ele an unc ional ca ego ies. The FunMap- pOne so wa e is open-sou ce and dis ibu ed unde he AGPL-3 license. A ailabili y and equi emen s P ojec name: FunMappOne P ojec home page: h ps://gi hub.com/G eco-Lab/ FunMappOne Ope a ing sys em(s): C oss-pla o m P og amming language: R O he equi emen s: Shiny License: AGPL-3 Scala e al. BMC Bioin o ma ics (2019) 20:79 Page 7 o 7 Any es ic ions o use by non-academics: Fo comme - cial use and modi ica ions please con ac he co espond- ing au ho . Addi ional iles Addi ional ile 1:FunMappOne use manual. Use manual o he FunMappOne ool. (DOCX 1940 kb) Addi ional ile 2:Excel ile con aining inpu da a o he case s udy. The excel ile is composed o one shee o each exposu e and a las shee con aining g ouping in o ma ion. Each exposu e shee is named wi h he exposu e ID and con ains wo columns con aining he lis o selec ed genes and he associa ed -s a is ics, espec i ely. The las shee con ains wo columns: one epo ing he lis o exposu e IDs and ano he he co esponding g oup. (XLSX 63 kb) Addi ional ile 3:Case s udy KEGG en ichmen maps. KEGG en ichmen maps showing modi ica ion di ec ion a e clus e ing analysis wi h 11 clus e s. Panel A ( op) shows en ichmen esul s summa ized a KEGG Le el 1, panel B (middle) shows en ichmen esul s summa ized a KEGG Le el 2, panel C (bo om) shows en ichmen esul s summa ized a KEGG Le el 3 (pa hways le el). (PPTX 6869 kb) Addi ional ile 4:Le el 1,2,3 Reac ome and Gene On ology (BP, CC, MF) maps o he p oposed case s udy. Reac ome maps ha e been p oduced by p o iding “Addi ional ile 1” as inpu and choosing “Reac ome” en ichmen , anno a ion was pe o med using “Bon e oni” as mul iple es ing co ec ion me hod wi h “0.001” as signi icance h eshold. Th ee classes o Gene On ology maps ha e been p oduced by p o iding “Addi ional ile 1” as inpu and choosing “GO” and al e na i ely “BP”, “CC” o “MF” en ichmen , anno a ion was pe o med using “Bon e oni” as mul iple es ing co ec ion me hod wi h “0.001” as signi icance h eshold. In bo h cases, o he plo ing “median” was chosen as summa y s a is ics and map colo s we e associa ed o he summa ized each e m modi ica ion di ec ion by choosing he sign op ion. (PDF 3044 kb) Abb e ia ions ARO: A oclo 1254; BP: Biological p ocesses; BPA: Bisphenol A; CA: Calyculin A; CC: Cellula componen s; CF: Clo ib a e; CSA: Cyclospo in A; CSPT: Cispla in; CT: Ca bon e achlo ide; DEM: Die hyl malea e; DIDP: Diisodecyl ph hala e; DM: D-manni ol; En .P: Te m en ichmen p- alue; ETP: E oposide; FK506: Tac olimus; HCE: Hep achlo epoxide; HCH: β-Hexachlo ocyclohexane; KEGG: Kyo o Encyclopedia o Genes and Genomes; LAC: Lead ace a e; MEN: Menadione; MF: Molecula unc ions; MMC: Mi omycin C; MNU: N-Me hyl-N-ni osou ea; MV: Gene modi ica ion alue; OA: Okadaic acid; PB: Phenoba bi al; SAR: Sodium a seni e; SS: Summa y s a is ics; TBTO: T ibu yl in oxide; TCE: 1,1,1,-T ichlo oe hane; TCDD: 2,3,7,8-Te achlo odibenzo-p-dioxin; TCPOBOP: 1,4-Bis[2-(3,5-dichlo opy idyloxy)]benzene; WY: Wye h-14643 Acknowledgemen s No applicable. Funding This s udy was suppo ed by he Academy o Finland (g an ag eemen s 275151 and 292307). A ailabili y o da a and ma e ials The FunMappOne ool, i s sou ce code and he example es da a used in his manusc ip a e a ailable a h ps://gi hub.com/G eco-Lab/FunMappOne. Au ho s’ con ibu ions GS and DG concei ed he applica ion and coo dina ed he p ojec . GS, AS and VSM de eloped he FunMappOne ool. GS, LAS, and DG analyzed and in e p e ed he esul s o he case s udy. GS, AS, VSM, DG and LAS d a ed he manusc ip . All au ho s ead and app o ed he inal manusc ip . E hics app o al and consen o pa icipa e No applicable. Consen o publica ion No applicable. Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. Au ho de ails 1Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, A o Ylpön ka u 34 - A o building, FI-33014 Tampe e, Finland. 2BioMediTech Ins i u e, Uni e si y o Tampe e, A o Ylpön ka u 34 - A o building, FI-33014 Tampe e, Finland. 3Ins i u e o Bio echnology, Uni e si y o Helsinki, Viikinkaa i 5d, FI-00014 Helsinki, Finland. Recei ed: 16 July 2018 Accep ed: 18 Janua y 2019 Re e ences 1. Huang DW, She man BT, Lempicki RA. Sys ema ic and in eg a i e analysis o la ge gene lis s using DAVID bioin o ma ics esou ces. Na P o oc. 2009;4(1):44–57. h ps://doi.o g/10.1038/np o .2008.211.9411012. 2. Kulesho MV, Jones MR, Rouilla d AD, Fe nandez NF, Duan Q, Wang Z, Kople S, Jenkins SL, Jagodnik KM, Lachmann A, McDe mo MG, Mon ei o CD, Gunde sen GW, Ma’ayan A. En ich : a comp ehensi e gene se en ichmen analysis web se e 2016 upda e. Nucleic Acids Res. 2016;44(Web Se e issue):90–7. h ps://doi.o g/10.1093/na /gkw377. 3. Chen J, Ba des EE, A onow BJ, Jegga AG. {ToppGene} {Sui e} o gene lis en ichmen analysis and candida e gene p io i iza ion. Nucleic Acids Res. 2009;37(Web Se e issue):305–11. h ps://doi.o g/10.1093/na /gkp427. 4. Reimand J, Kull M, Pe e son H, Hansen J, Vilo J. g:{P o ile }—a web-based oolse o unc ional p o iling o gene lis s om la ge-scale expe imen s. Nucleic Acids Res. 2007;35(Web Se e issue):193–200. h ps://doi.o g/10.1093/na /gkm226. 5. Yu G, Wang L-G, Han Y, He Q-Y. {clus e P o ile }: an {R} {Package} o {Compa ing} {Biological} {Themes} {Among} {Gene} {Clus e s}. OMICS : J In eg Biol. 2012;16(5):284–7. h ps://doi.o g/10.1089/omi.2011.0118. 6. Wal e W, Sánchez-Cabo F, Rico e M. {GOplo }: an {R} package o isually combining exp ession da a wi h unc ional analysis. Bioin o ma (Ox o d, England). 2015;31(17):2912–4. h ps://doi.o g/10.1093/bioin o ma ics/ b 300. 7. Fo ino V, Alenius H, G eco D. {BACA}: bubble {chA } o compa e anno a ions. BMC Bioin o ma ics. 2015;16(1). h ps://doi.o g/10.1186/ s12859-015-0477-4. 8. Kanehisa M, Sa o Y, Kawashima M, Fu umichi M, Tanabe M. KEGG as a e e ence esou ce o gene and p o ein anno a ion. Nucleic Acids Res. 2016;44(D1):457–62. h ps://doi.o g/10.1093/na /gk 1070. 9. Fab ega A, Jupe S, Ma hews L, Sidi opoulos K, Gillespie M, Ga apa i P, Haw R, Jassal B, Ko ninge F, May B, Milacic M, Roca CD, Ro h els K, Se illa C, Shamo sky V, Sho se S, Va usai T, Vi e i G, Weise J, Wu G, S ein L, He mjakob H, D’Eus achio P. The Reac ome Pa hway Knowledgebase. Nucleic Acids Res. 2018;46(D1):649–55. 10.1093/na /gkx1132.NIHMS150003. 10. Ashbu ne M, Ball CA, Blake JA, Bo s ein D, Bu le H, Che y JM, Da is AP, Dolinski K, Dwigh SS, Eppig JT, Ha is MA, Hill DP, Issel-Ta e L, Kasa skis A, Lewis S, Ma ese JC, Richa dson JE, Ringwald M, Rubin GM, She lock G. Gene on ology: Tool o he uni ica ion o biology. 2000. 10614036.h ps://doi.o g/10.1038/75556.h p://www.ncbi.nlm.nih.go / pubmed/10802651.h p://www.pubmedcen al.nih.go /a icle ende . cgi?a id=PMC3037419.h p://www.na u e.com/doi inde /10.1038/ 75556. 11. Schaap MM, Wacke s PFK, Zwa EP, Huijskens I, Jonke MJ, Hend iks G, B ei TM, an S eeg H, an de Wa e B, Luij en M. A no el oxicogenomics-based app oach o ca ego ize (non-)geno oxic ca cinogens. A ch Toxicol. 2015;89(12):2413–27. h ps://doi.o g/10.1007/ s00204-014-1368-6.