Golden rules for building sustainable bioinformatics capacity using Nextflow and nf-core with a focus on early-mid career researchers: The Kids Research Institute Australia, case report.

Golden ules o building sus ainable bioin o ma ics capaci y using 1 Nex low and n -co e wi h a ocus on ea ly-mid ca ee esea che s: The 2 Kids Resea ch Ins i u e Aus alia, case epo . 3 Pa icia Agudelo-Rome o1,2,3,*,#, Talya Con adie1, Jose A. Capa os-Ma in1,4,5, Da id J. 4 Ma ino1, An hony Kicic1,6,7, S ephen M. S ick7,8, Ch is ophe Hakkaa 9, Abhina Sha ma10,#, 5 and he Theme Collabo a ion G oup. 6 7 1Wal-Yan Respi a o y Resea ch Cen e, The Kids Resea ch Ins i u e Aus alia, Pe h, Wes e n 8 Aus alia, Aus alia. 9 2Aus alian Resea ch Council Cen e o Excellence in Plan Ene gy Biology, School o Molecula 10 Sciences, The Uni e si y o Wes e n Aus alia, Pe h, Wes e n Aus alia, Aus alia. 11 3Eu opean Vi us Bioin o ma ics Cen e , F ied ich-Schille -Uni e si a Jena, Thu ingia, Ge many. 12 4Cu in Heal h Inno a ion Resea ch Ins i u e (CHIRI), Cu in Uni e si y, Pe h, Wes e n Aus alia, 13 Aus alia. 14 5School o Medicine, The Uni e si y o Wes e n Aus alia, Pe h, Wes e n Aus alia, Aus alia. 15 6School o Popula ion Heal h, Cu in Uni e si y, Pe h, Wes e n Aus alia, Aus alia 16 7Cen e o Cell The apy and Regene a i e Medicine, Medical School, The Uni e si y o Wes e n 17 Aus alia, Pe h, Wes e n Aus alia, Aus alia. 18 8Depa men o Respi a o y and Sleep Medicine, Pe h Child en’s Hospi al, Pe h, Wes e n Aus alia, 19 Aus alia. 20 9Seqe a Labs, Ba celona, Ca alonia, Spain. 21 10DSI-NRF Cen e o Excellence o Biomedical Tube culosis Resea ch; SAMRC Cen e o 22 Tube culosis Resea ch; Di ision o Molecula Biology and Human Gene ics, Facul y o Medicine and 23 Heal h Sciences, S ellenbosch Uni e si y, Cape Town, Wes e n Cape, Sou h A ica. 24 * Co espondence: 25 Co esponding Au ho : [email p o ec ed] 26 # Co-senio au ho ship 27 Keywo ds: Bioin o ma ics, Capaci y Building, Nex low, Ea ly-Mid Ca ee Resea che , Omics, 28 Pipelines 29 Running Ti le: Building sus ainable bioin o ma ics capaci y using Nex low and n -co e 30 31 32 33 34 35 THIS PREPRINT WAS IMPROVED FURTHER AND THE FINAL FORM IS NOW PUBLISHED IN FRONTIERS IN BIOINFORMATICS 2 Abs ac 36 The inc easing adop ion o high- h oughpu "omics" echnologies has heigh ened he demand 37 o s anda dized, scalable, and ep oducible bioin o ma ics wo k lows. Nex low and n -co e p o ide 38 a obus amewo k o esea che s, pa icula ly ea ly- and mid-ca ee esea che s (EMCRs), o 39 na iga e complex da a analysis. A The Kids Resea ch Ins i u e Aus alia, we implemen ed a s uc u ed 40 app oach o bioin o ma ics capaci y building using hese ools. This pe spec i e p esen s nine p ac ical 41 “golden” ules ha acili a ed he success ul adop ion o Nex low and n -co e, add essing 42 implemen a ion, knowledge gaps, esou ce alloca ion, and communi y suppo . Ou expe ience se es 43 as a guide o ins i u ions aiming o es ablish sus ainable bioin o ma ics capabili ies and empowe 44 EMCRs. 45 46 In oduc ion 47 The demand o esea che s skilled in bioin o ma ics has su ged due o he inc easing adop ion 48 o high- h oughpu “omics” echnologies, p esen ing bo h, oppo uni ies and challenges. Resea che s 49 wi hou o mal aining in p og amming o wo k low engines mus now analyze high-dimensional 50 da ase s using a ious ools, while ensu ing analysis a e ep oducible ac oss compu a ional 51 en i onmen s. This has highligh ed a need o omics da a wo k lows ha a e s anda dized, scalable, 52 and ep oducible. In his con ex , we aimed o p o ide ea ly- and mid-ca ee esea che s (EMCRs) wi h 53 oppo uni ies o gain knowledge and es ablish a bioin o ma ics suppo sys em (Woelme e al., 2021) 54 using wo k low manage s like Nex low and he n -co e communi y. 55 Nex low is a wo k low managemen sys em designed o s eamline bioin o ma ics esea ch by 56 composing mul iple ools in o a single pipeline (Di Tommaso e al., 2017). I s lexibili y, po abili y 57 and scalabili y anges om local se e s, high-pe o mance clus e s, and cloud en i onmen s wi h 58 minimal econ igu a ion. Nex lows na i e ask pa alleliza ion, makes i ideal o handling high-59 olume o da ase s o igina ing om mode n "omics" echnologies (Dai and Shen, 2022). In addi ion, 60 he n -co e communi y enhances he Nex low ecosys em by connec ing use s h ough hacka hons, 61 semina s, aining, and collabo a i e pla o ms like Slack (Ewels e al., 2020; Lange e al., 2024). Fo 62 newcome s, his suppo is in aluable o oubleshoo ing and p o essional g ow h. Addi ionally, n -63 co e communi y de elops and main ains pipelines used ac oss di e en ields o li e sciences. Each 64 pipeline ollows s ic guidelines o ensu e obus ness, ep oducibili y, and ease o use, which a e key 65 ea u es o EMCRs en e ing bioin o ma ics. 66 THIS PREPRINT WAS IMPROVED FURTHER AND THE FINAL FORM IS NOW PUBLISHED IN FRONTIERS IN BIOINFORMATICS 3 Fo ins i u ions and esea che s in "omics" s udies, adop ing Nex low and n -co e is s a egic. 67 These ools enable managemen o high- h oughpu da a, e ec i e collabo a ion, and ep oducible 68 esul s, ensu ing con inued scien i ic ad ancemen . This is especially c ucial o EMCRs d i ing 69 inno a ion. By in eg a ing hese ools in o aining, ins i u ions can be e p epa e esea che s o mee 70 mode n bioin o ma ics challenges (S ephens e al., 2015). 71 He e, we sha e ou expe ience a The Kids Resea ch Ins i u e Aus alia (The Kids), a esea ch 72 ins i u ion in Pe h (Aus alia) wi h a di e se wo k o ce and a s ong commi men o building EMCR 73 expe ise in compu a ional biology. The Theme Collabo a ion G oup, a mul idisciplina y eam ac oss 74 The Kids, played a key ole in shaping his ini ia i e (Supplemen al Da a S1). We ou line p ac ical 75 “golden” ules essen ial o adop ing and implemen ing Nex low (Table 1). These guidelines aim o 76 help o he ins i u ions es ablish sus ainable and obus bioin o ma ics capabili ies, enabling esea che s 77 o ully le e age Nex low and n -co e o scien i ic disco e y. Figu e 1 summa izes he key ules and 78 he imeline o ou ine e en s c i ical o Nex low adop ion. 79 80 Rule 1: De elop a s a egic plan aligned o he ins i u ional p io i ies 81 A well-designed capaci y de elopmen p og am wi hin any ins i u e should (i) e lec he needs 82 o he ins i u ion and (ii) adhe e o a well-scoped cu iculum ha bene i s he en i e o ganiza ion. 83 Iden i ying co e s akeholde s and aligning he p og am wi h he ins i u ion's cu en and u u e da a 84 analysis needs is essen ial o ensu ing i s long- e m ele ance and impac (A on e al., 2021). 85 A The Kids, ou ision is "Happy Heal hy KIDS," and ou objec i e is o enhance he heal h, 86 de elopmen , and li es o child en and adolescen s h ough exempla y esea ch. Ou wo k 87 encompasses a wide a ay o disciplines, om undamen al o applied science. In ligh o his di e si y, 88 we ha e concen a ed on esea ch s eams dependen on omics da a p ocessing, p o iding aining o 89 s uden s and EMCRs engaged in hese da a-in ensi e ields. We hos ed an in e nal own hall mee ing, 90 a ended by esea ch eams engaged in bioin o ma ics, as well as membe s o he Pawsey 91 Supe compu ing Resea ch Cen e, an Aus alian go e nmen -suppo ed high-pe o mance compu ing 92 (HPC) acili y. Du ing his mee ing, we documen ed common oadblocks and challenges expe ienced 93 by bioin o ma ic use s, which laid he ounda ion o de eloping a s a egic plan and he o ma ion o 94 a special in e es g oup (SIG) wi h ocus on bioin o ma ics. Building on hese insigh s, we es ablished 95 a h ee-pilla s a egy aimed a imp o ing he e iciency, epea abili y, and scalabili y o ou 96 bioin o ma ics in es iga ions. This s a egy ocuses on ansi ioning om ad-hoc shell sc ip s o he 97 4 in eg a ion o Nex low wo k low manage in o ou compu a ional esea ch (Sz uka e al., 2024) while 98 making he bes use o sha ed esou ces. 99 1. Pilla one o ou s a egies in ol ed iden i ying a ailable compu a ional esou ces, 100 unde s anding hei alloca ion ac oss he di e en esea ch g oups, and assessing exis ing 101 knowledge gaps and bioin o ma ics needs. 102 2. Pilla wo ocused on p omo ing a communi y o “powe use s” by in eg a ing suppo and 103 men o ship. A men o ship p og am was es ablished o guide EMCRs h ough hei lea ning 104 jou ney, while p omo ing collabo a ion, a co ne s one o da a-d i en esea ch. Resea che s 105 we e encou aged o sha e hei expe iences, common challenges, and bes p ac ices, 106 ul ima ely s eng hening knowledge exchange, and spa king inno a ion wi h a s ong, 107 suppo i e communi y. 108 3. Pilla h ee emphasized adhe ence o es ablished s anda ds and bes p ac ices. We 109 p omo ed he use o bioin o ma ics s anda ds and p o ided empla es and guidelines o 110 c ea ing ep oducible and well-documen ed Nex low pipelines, he eby enhancing esea ch 111 quali y and c edibili y. 112 These h ee pilla s in o med he design o engagemen ac i i ies, including semina s and online 113 u o ials (Rule 5), Hacky-Hou s (Rule 6), one-on-one sessions and boo camps (Rule 7). Addi ionally, 114 con inuous e alua ion and eedback (Rule 8 and 9) we e pi o al in e ining and adap ing hese ac i i ies 115 o mee he e ol ing needs o esea che s e ec i ely. 116 Rule 2: P omo e awa eness o ep oducible, obus and open science. 117 P omo ing awa eness o ep oducible da a analysis and bioin o ma ics bes p ac ices is c ucial 118 in “omics” ields like ansc ip omics, genomics, p o eomics, me abolomics, and mic obiomics. As 119 esea che s inc easingly adop “mul i-omics” app oaches o gain deepe insigh s in o biological 120 phenomenon (Reel e al., 2021; Vahabi and Michailidis, 2022), he impo ance o ep oducibili y, 121 obus ness and open science canno be o e s a ed. These p ac ices ensu e eliable and ac ionable 122 esea ch indings while p omo ing anspa ency and allowing o alida ion, which is i al o scien i ic 123 p og ess (K aus, 2014; S ewa e al., 2022). A success ul implemen a ion o he capaci y-building 124 s a egy (Rule 1) should ensu e clea communica ion o he bes bioin o ma ic p ac ices. Pa icipan s 125 mus unde s and he impo ance o hese p ac ices in p og amming and compu a ional expe imen s, 126 such as es ing simula ions, algo i hms, and models. Addi ionally, hey should ecognize he clea 127 5 bene i s o applying hese p ac ices wi hin he con ex o hei own wo k and esea ch g oups. (Ca ey 128 and Papin, 2018; Heise e al., 2023). 129 A The Kids, we accomplished his h ough (i) di ec app oaches, using semina s and Hacky-130 Hou s, as well as (ii) indi ec ly h ough mailing lis s and a dedica ed Mic oso Teams channel. These 131 pla o ms allowed us o sha e ele an esou ces and p omo e awa eness e ec i ely. 132 Rule 3: Iden i y he needs and skill gaps o di e en esea ch g oups wi hin he ins i u e 133 The key o success ul capaci y building e o lies in unde s anding exis ing skills and 134 iden i ying u u e needs o esea ch g oups (Woelme e al., 2021). We began by conduc ing an ini ial 135 su ey aimed a comp ehensi ely e alua ing key equi emen s o he EMCRs (Supplemen al Da a S2). 136 The su ey ga he ed in o ma ion on (i) basic compu a ional skills (e.g. Linux command line 137 p o iciency o sc ip ing using Py hon and/o R languages), (ii) amilia i y wi h bioin o ma ics ools 138 and wo k lows; (iii) expe ience wi h Nex low and n -co e; (i ) and speci ic challenges aced in hei 139 esea ch p og ams. 140 We encou aged pa icipa ion by emphasizing he impo ance o he su eys in iden i ying 141 bioin o ma ic pipeline needs a The Kids, wi h weekly ollow-ups o e ou weeks. This app oach 142 ensu ed de ailed and hones esponses. The esul s p o ided an accu a e o e iew o he cu en 143 bioin o ma ics capabili ies and needs, and he insigh s shaped ou implemen a ion s a egy, helping us 144 iden i y p io i y opics o aining and skill de elopmen . 145 Using insigh s om he su ey, we ca ego ized esea che s’ skills in o h ee le els (beginne , 146 in e media e, and ad anced). This allowed us o design a ge ed aining p og ams o add ess he 147 iden i ied gaps by building upon he sensi ize, ain, hack and collabo a e model (Ka ega e al., 2023). 148 1. Beginne use s ocused on basic skills and a ended in oduc o y cou ses on Linux, 149 bioin o ma ics undamen als (B andies and Hogg, 2021), and moni o ing pipelines using 150 he Seqe a Pla o m (h ps://seqe a.io/pla o m/). 151 2. In e media e use s ocused on wo k low de elopmen wi h Nex low and n -co e pipeline 152 empla es (Roach e al., 2022). 153 3. Ad anced use s specialized in cus omizing exis ing n -co e pipelines and in eg a ing new 154 ools o clus e sys ems, such as hose p o ided wi hin Aus alia by Aus alian Pawsey 155 Supe compu ing Resea ch Cen e (Pawsey Supe compu ing Resea ch Cen e Pe h, 2023d, 156 6 2023b, 2023c, 2023a) and he Aus alian BioCommons Leade ship Sha e (ABLeS) 157 p og am (Gus a sson e al., 2023). 158 This a ge ed app oach ensu ed ha esea che s ecei ed he p ecise suppo hey needed o ad ance 159 hei bioin o ma ics skills and p og ess e icien ly wi hin hei esea ch p ojec s. 160 Rule 4: Engage wi h he IT eam o in as uc u e au oma ion and sus ainabili y h ough 161 documen a ion (Nex low-Biowiki). 162 Building a sus ainable bioin o ma ics en i onmen equi es close collabo a ion wi h he 163 In o ma ion Technology (IT) eam. Thei expe ise in in as uc u e managemen , da a s o age (sho 164 and long- e m), and da a sha ing and p o ec ion is in aluable o adhe ing o esponsible big da a 165 esea ch p ac ices (Zook e al., 2017). A The Kids, we pa ne ed wi h he IT eam o s anda dize and 166 s eamline equen compu a ional eques s o “omics” da a analysis and o es ablish long- e m s o age 167 solu ions o he Nex low-Biowiki documen a ion c ea ed du ing he p og am (Agudelo-Rome o e 168 al., 2023). 169 To op imize in as uc u e, he IT eam de eloped a baseline i ual machine (VM) empla e. 170 This empla e s anda dized con igu a ions o e icien ins alla ion o he co e dependencies, including 171 Ja a (LTS e sion) (A nold e al., 2005), Nex low (Di Tommaso e al., 2017), and n -co e (Ewels e 172 al., 2020). Package manage s like Bioconda (Dale e al., 2018) and Mamba (mamba-O g, n.d.), and 173 con aine iza ion ools such as Docke (da Veiga Lep e os e al., 2017) and Singula i y (Ku ze e al., 174 2017), we e also p econ igu ed. Sha ed access o c i ical esou ces (e.g. genome e e ence iles) was 175 enabled, and compu ing esou ces, such as numbe o CPU co e, memo y and disk space (sc a ch 176 wo kspaces), we e alloca ed based on p ojec -speci ic da a and analysis needs. To ensu e obus ness, 177 we alida ed he se up by unning a ious n -co e pipelines, add essing any issues, ensu ing a seamless 178 pe o mance, and secu ing enough s o age o he ou come’s iles on he VMs. 179 A he same ime, o ensu e long- e m sus ainabili y, we c ea ed comp ehensi e documen a ion 180 ha could be used o u u e e e ence o EMCR and esea che s’ communi y. We hos ed he 181 Nex low-BioWiki p ojec (Agudelo-Rome o e al., 2023) on The Kids’ ins i u ional Gi Hub eposi o y 182 (h ps://gi hub.com/Tele honKids). This esou ce includes de ailed guides on bioin o ma ics ools, 183 wo k lows execu ion, oubleshoo ing in ou in as uc u e, online manuals, ideo u o ials, and 184 F equen ly Asked Ques ions (FAQs); ca e ing o di e se lea ning s yles. The documen a ion was 185 designed o be use - iendly and accessible o all esea che s. Regula upda es based on use s’ eedback 186 THIS PREPRINT WAS IMPROVED FURTHER AND THE FINAL FORM IS NOW PUBLISHED IN FRONTIERS IN BIOINFORMATICS 7 ha e ensu ed ha esou ces emain ele an and aligned wi h echnological ad ancemen s and 187 communi y needs. 188 Rule 5: Conduc egula semina s using ele an n -co e pipelines. 189 Regula semina s aimed o in oduce pa icipan s o conduc ing compu a ional expe imen s 190 unde op imal condi ions (“ he happy pa h”), ocusing on da a analysis and in e p e a ion. Fo he 191 implemen a ion o mon hly semina s, we ollowed p inciples ou lined by Fadlelmola e al 2019. 192 Semina s we e designed o add ess eal-wo ld da a analysis challenges, such as pa ame e op imiza ion 193 and chaining mul iple pipelines o comp ehensi e analyses. 194 A The Kids, egula semina s co e ed he b ead h o a ailable esou ces, as key e e ences, 195 wi hou o e whelming pa icipan s. These egula semina s also o e ed a unique oppo uni y o 196 inc ease awa eness abou he esou ces a ailable om he Ins i u e and he Nex low and n -co e 197 communi ies, such as pipeline-speci ic “by e size” alks. To acili a e ocused lea ning, we o ganized 198 dedica ed semina s o “omics” pipelines unde cohesi e hemes, including: 199 1. T ansc ip omics: n -co e/ naseq (Pa el e al., 2024), n -co e/di e en ialabundance 200 (Wacke O e al., 2023), n -co e/sm naseq (Pel ze e al., 2024b). 201 2. Me hyla ion: n -co e/me hylseq (Ewels e al., 2024) , n -co e/a acseq (Pa el e al., 2023). 202 3. Mic obiome: n -co e/ampliseq (S aub e al., 2024) , n -co e/mag (Ya es e al., 2024), 203 agudelo ome o/e e es _n (Agudelo-Rome o e al., 2025). 204 4. Single-cell: n -co e/sc naseq (Pel ze e al., 2024a). 205 This hema ic g ouping allowed pa icipan s o di e deepe in o speci ic a eas, es ablishing a 206 be e unde s anding o key biological concep s ela ed o he “omics” s a egies, while also 207 encou aging collabo a ions among s uden s and EMCRs wo king on simila opics. 208 Ul ima ely, du ing ou semina s, we encou aged pa icipan s o explo e and emb ace cloud 209 pla o ms as an in eg al pa o hei con inued g ow h. O e ing auxilia y oppo uni ies o alen 210 enhancemen and pe o ming hei da a analysis in he cloud. Consequen ly, ce ain pa icipan s in his 211 p og am ob ained esea ch c edi s o pe o ming hei analysis and deploying Nex low on cloud 212 in as uc u e. Supplemen al Table S1 delinea es cloud compu ing companies ha ex end c edi s o 213 academic pu poses. 214 215 8 Rule 6: O ganize p ac ical aining sessions (Hacky-Hou s) 216 While egula semina s in oduce heo e ical concep s, Hacky-Hou s ansla e hem in o 217 p ac ical skills h ough coo dina ed sessions ailo ed o he speci ic needs o EMCRs (Heise e al., 218 2023). These sessions p o ide hands-on oppo uni ies o add ess challenges such as accessing di e en 219 in as uc u es o op imizing compu ing sou ces o di e se p ojec s, empowe ing EMCRs o build 220 hei analy ical capaci y e ec i ely. 221 The in ol emen o he IT eam and esea che s in designing Hacky-Hou s p o ided aluable 222 insigh in he sessions. The IT eam con ibu ed by o e ing guidance on s uc u ed use o esou ces, 223 ensu ing e icien esou ce alloca ion. Meanwhile, engaging wi h esea che s helped iden i y 224 oppo uni ies o e ine he p og am’s con en . Fo ins ance, when a speci ic n -co e pipeline was 225 ou inely used, an in e ac i e Hacky-Hou session was o ganized on ha pipeline. This session 226 included s ep-by-s ep ins uc ions and oubleshoo ing, deli e ing p ac ical and a ge ed aining 227 (Boudabous and Tekaia, 2020). 228 A The Kids, bi-weekly Hacky-Hou s (30-90 minu es) we e s uc u ed a ound opics iden i ied 229 ia pa icipan su eys. Ini ially, hese sessions ocused on undamen al skills such as downloading 230 and con igu ing n -co e pipelines. As Nex low and n -co e adop ion g ew, su eys e ealed a g owing 231 need o ad anced opics like oubleshoo ing and handling complex con igu a ions (Supplemen al 232 Da a S2). This i e a i e eedback p ocess allowed us o adap and p io i ize u u e session, ensu ing 233 ele ance o he e ol ing needs o he communi y. Hacky-Hou s we e scheduled lexibly and only held 234 when he e was clea in e es , maximizing pa icipa ion and a oiding unnecessa y s ain on 235 communica ion channels. 236 Rule 7: S uc u ed and Pe sonalized T aining 237 To accommoda e di e se skill le els and lea ning s yles, a success ul capaci y-building 238 p og am should combine s uc u ed g oup aining wi h pe sonalized men o ing, such as boo camps 239 and men o ing session, espec i ely. Boo camps a e an e icien way o imme se pa icipan s in he 240 heo e ical and p ac ical aspec s, exposing hem o he ools and echniques essen ial o mode n 241 bioin o ma ics. While men o ing sessions is pe sonalized app oach empowe ed indi iduals o na iga e 242 he complexi ies o bioin o ma ics. 243 A The Kids, we implemen ed an in eg a ed app oach ha included imme si e week-long 244 boo camps alongside ailo ed one- o-one men o ship sessions. Boo camps e ec i ely in oduced 245 9 pa icipan s, especially s uden s and EMCRs, o he heo e ical and p ac ical aspec s o Nex low and 246 n -co e. These e en s balanced concep ual lea ning wi h hands-on pipeline de elopmen and included 247 dedica ed "B ing You Own Da a" (BYO-D) sessions o ensu e ele ance o indi idual p ojec s. 248 Including IT eam membe s in hese sessions os e ed mu ual unde s anding be ween esea che s and 249 echnical suppo , aligning in as uc u e solu ions wi h esea ch needs. 250 To complemen g oup aining, we o e ed one- o-one men o ing ailo ed o pa icipan s wi h 251 limi ed bioin o ma ics expe ience o speci ic p ojec challenges. Men o s om he o ganizing eam 252 p o ided indi idualized suppo , helping pa icipan s gain con idence in using wo k low ools. These 253 sessions also gene a ed aluable eedback, which shaped and expanded he BioWiki documen a ion 254 (h ps://gi hub.com/Tele honKids/Nex low-BioWiki), ensu ing i emained p ac ical and use -255 in o med (McG a h e al., 2019). Fo b oade each, we ecommend s a ing wi h ounda ional semina s 256 (Rule 5) and Hacky-Hou s (Rule 6) be o e launching ad anced boo camps ocused on oubleshoo ing 257 and complex con igu a ions (Hagan e al., 2020). This laye ed app oach p omo es con inuous lea ning 258 and maximizes long- e m impac . 259 Rule 8: Facili a e con inuous lea ning and engagemen 260 To ensu e he long- e m success o he capaci y de elopmen p og am and build a communi y 261 o “powe -use s,” i 's c ucial o es ablish con inual lea ning and engagemen oppo uni ies. Le e aging 262 in e nal pla o ms, such as messaging apps o newsle e s, is an e ec i e s a egy o in o ming 263 pa icipan s abou cu en ac i i ies and upcoming e en s. A mon hly newsle e is also an e ec i e 264 medium o sha ing upda es on Nex low de elopmen s, like module bina ies and wa e con aine s, and 265 n -co e communi y p ojec s, including new pipelines, modules, and con igu a ions. 266 A The Kids, we es ablished a Mic oso Teams channel as a space o discussion and p omo ed 267 a local olun ee communi y. The long- e m success o he p og am depends on use s willing o assis 268 each o he in a eas such as c ea ing new so wa e/pipelines (B ack e al., 2022), cus omizing exis ing 269 pipelines, add essing in as uc u e ques ions, and designing bioin o ma ics expe imen s wi hin he 270 in as uc u e. 271 Beyond he in e nal communica ion, engagemen wi h he global Nex low and n -co e 272 communi ies is also encou aged and should be a majo goal o he capaci y-building p og am o long-273 e m success. Communi ies such as online o ums o e aluable esou ces, his is demons a ed in he 274 n -co e Slack g oup (h ps://n -co. e/join) and he o ganized communi y e en s, including online 275 aining sessions (h ps://n -co. e/e en s/hacka hon). By pa icipa ing in hese communi y e en s, 276 16 Table 1. Summa y o he mo i a ion o he ules’ de elopmen . 454 Mo i a ion Rule Implemen a ion Rule 1 and 2 Knowledge gap Rule 3 Resou ces Rule 4 Suppo Rule 5, 6, 7 and 8 Engagemen and e alua ion Rule 9 455 456 17 Au ho Con ibu ions 457 458 PAR: Concep ualiza ion, Funding acquisi ion, Resou ces, Supe ision, W i ing (o iginal d a , e iew 459 & edi ing). TC: Resou ces, W i ing ( e iew & edi ing). JAC: Concep ualiza ion, Funding acquisi ion, 460 W i ing (o iginal d a , e iew & edi ing). DJM: Funding acquisi ion, W i ing ( e iew & edi ing). AK: 461 Funding acquisi ion, W i ing ( e iew & edi ing). SMS: Funding acquisi ion, W i ing ( e iew & 462 edi ing). CH: Resou ces, W i ing ( e iew & edi ing). AS: Concep ualiza ion, Resou ces, W i ing 463 (o iginal d a , e iew & edi ing). 464 Con lic o In e es 465 The au ho s decla e ha he esea ch was conduc ed in he absence o any comme cial o inancial 466 ela ionships ha could be cons ued as a po en ial con lic o in e es . 467 468 Funding 469 This p ojec was unded by The Kids Resea ch Ins i u e Aus alia's Theme Collabo a ion Awa d and a 470 Google Cloud Educa ion P og am g an . 471 Acknowledgmen s 472 We would like o hank he key con ibu ions o The Kids IT eam ha pa icipa ed in he de elopmen 473 o his p ojec , S e en Figliomeni, Kie an Gee and Ch is Hadden. 474