biostudies-literatureGrapotte MNCI NIH HHSWellcome Trust3297https://www.ebi.ac.uk/biostudies/studies/S-EPMC8172540biostudies-literatureUnknown12(1)Using the Cap Analysis of Gene Expression (CAGE) technology, the FANTOM5 consortium provided one of the most comprehensive maps of transcription start sites (TSSs) in several species. Strikingly, ~72% of them could not be assigned to a specific gene and initiate at unconventional regions, outside promoters or enhancers. Here, we probe these unassigned TSSs and show that, in all species studied, a significant fraction of CAGE peaks initiate at microsatellites, also called short tandem repeats (STRs). To confirm this transcription, we develop Cap Trap RNA-seq, a technology which combines cap trapping and long read MinION sequencing. We train sequence-based deep learning models able to predict CAGE signal at STRs with high accuracy. These models unveil the importance of STR surrounding sequenNature communicationsDiscovery of widespread transcription initiation at microsatellites predictable by sequence-based deep neural network.PMC8172540P30 CA008748106954/Z/15/ZU01 CA151118Bessiere CBickel PJSese JSaito MKaminuma EKawaguchi SMatsuda HWinteringham LNYumoto NForrester LMWatanabe KBeltrami CAKajiyama KRost BIida KBusch ALilje BKawai JOhshima MZucchelli SBrombacher FKawamura YIKuznetsov VAFrith MCIkeo KYoung RYu NYFarach-Carson MCSaito AFaghihi MSeno SHerlyn MBult CJFreeman TCBajic VBIvshina AVHugues RPonten FMotakis ELopez GHorie MRehli MHashimoto KHale PSakaue SKratz ALipovich LSoderhall CMungall CJCoskun MMinoda AFunayama MKitano HTesta ACEndoh MBansal MSato HMak GKPachkov MBertin NOgishima SFurlanello CKlinken SPNotredame CGhosheh YBoyd MMikami NMurakawa YTaylor MSSachenkova OChen YPoulsen TMDalla ESuzuki AMCameron CJThongjuea SKojima SPassier RMagi SNakamura KLassmann TKojima MKundaje AKunikata HMedvedeva YAYasuzawa KKulakovskiy INakamura YEdinger MSchafferhans AItoh Mde Hoon MJLSemple CAGeijtenbeek TBHZhang PGHitchens KJDi Bernardo DKlepper KAsai KJohnson RHou RAstrom GRye MBCarbone AYoneda MTaguchi ADiehl ADNishiguchi KMAbugessaisa IChierici MWatanabe SIshizu YSjostedt EMandai MRoudnicky FGustincich SKogo YVelmeshev DClark ELMenichelli CHorimoto KRavasi TShimoni Yvan Nimwegen ESchueler MFurusawa CGeier FKoseki HTerada ATagami MKanaya TAlasiri RAlam TCaglio GHandoko LMorishita SChristodoulou JDaub COAlam IOhtsu TCalifano ANguyen QHMathelier AForsberg MBonetti AGusev OHamada MNoma SSaetrom PEhrlund AMason EFerrai CBrehelin LTarui HThomas MSchneider CYoshihara MKai CSaraswat MBlanchette MGough JDavis CAPersson HSimonSanchez JKomiyama HYoshida EKawashima TGreco DDjebali SBodega BKoyasu SPhilippens IHMiyagi YTsang HGSeverin JBaker SBono HAitken SYoshida YZabierowski SELaw ABaillie JKMar JVitting-Seerup KFukuda SMeehan TFMorimoto MGao HGuhl SEnomoto SEnomoto HKaczkowski BPradhan SIkawa TJorjani HSai HTanzer ATsujikawa MNishiyori-Sueki HLarouche MPridans CArchibald ALManabe RToyoda HNason GPSakai MBatzoglou SPlessy CTakamochi KKapranov PKomatsu NFANTOM consortiumYano KLizio MYokokura SYo MBeltrami APHide WDenisenko EOksvold PChen JCSheng GVitezic MHasegawa AMurata MFuruno MKasukawa TTakahashi MGruber AJPark SJTakahashi HMiyawaki ASchmeier SWolvetang EBhattacharya SKiryu HPiazza SWahlestedt CValen EKenna TJYatsuka YSeo WHatano TAlhendi AMNMakeev VArakawa TSchmocker DCarter KWMaruyama TMummery CLQin XYRamilowski JGuler RBlake JACesselli DMorioka MVoolstra CRJankovic BRUhlen MInoue NYamada TTerao YSakaguchi SBougouffa SSandelin ASano HBrown JBKawasaki SStoiber MHAlinejad-Rokny HNepal CNagase TRoy RTomii KHallstrom BMMoro KArbel TMukarram AKArner PMejhert NSaiki SYoung RSFavorov AFrancescatto MOkada-Hatakeyama MKishima MGhosh SArner EDohi TDalby MArcher JSugiyama DOw GSEdge ASBTakenaka YSakaue-Sawano AFaulkner GJDrablos FBttcher MKwon TJSivertsson ADeviatiiarov RLenhard BTanaka HLe-Cao KASaxena AKawaji HLi RLi KSummers KMSui SHBeckhouse AGWasserman WWCarninci PCiani YSwoboda RKPaparountas TAraki MBurroughs AMSuzuki TShin JWSuzuki MAshoor HSuzuki NPrezioso CRennie SSuzuki KHayashi RSuzuki HMuramatsu MOkazaki YRoos MYagi KLennartsson Avon Feilitzen KPascarella GDostie JRackham OTam KTFagiolini MNishida KCannistraci CVHase TMaruyama KNakai KKojima-Ishiyama MKloppmann ECarlisle AJHa TJOhkura NTaniuchi IKasianov ASKitajima HNakahara FMorikawa HIenasescu HKlaric Ede Rie DAndersson RBalic ABajpai ARito TYamanaka YElofsson AQuackenbush JKaneda HNitta KBatagov AOForrest ARRDetmar MEkwall KChang JCHarrow JHorton POzturk MKuo RMiura HLecellier CHKhachigian LHarshbarger JKukalev AKatayama TTang DKuo TTanaka YFernandes AMTanaka RBaldarelli RMYabukami HGoldowitz DGrapotte MSchulze-Tanzil GHara MFrith MKnox AJHori FNielsen LKDieterich LCGazova IKono TGingeras TRIkemura THuang YKaida KChatelain CLaurent GSDimont EShi JBreda JLee Wvan de Wetering MHuang ENakachi YBabina MIto YFort AGebhard CGoldberg TKuraku SNagasaka KOhno HGojobori TYamamoto MVorontsov IEHeutink PSasamoto YVerardo RNakazawa TMeynert ANishiyori HZwahlen MSaya HSwanson DJRamilowski JASchultes EAKaliyaperumal RKodama YRobert CAlvarez MJYamamoto YHarbers M't Hoen PACNoguchi SHaberle VHayashizaki YKempfle JSYamaguchi YGuigo RMizuno YHon CCPoulain SParihar SPWells CAPombo AKere JJenjaroenpun PAken BLBurt DWTakeda SKawamoto HHume DDeplancke BJoshi AKato SBornholdt JTorlai Triglia EJorgensen MRizzu POkada YKitamura THattori NJurman GMotohashi HSchmidl COrlando VfalseDiscovery of widespread transcription initiation at microsatellites predictable by sequence-based deep neural network.Using the Cap Analysis of Gene Expression (CAGE) technology, the FANTOM5 consortium provided one of the most comprehensive maps of transcription start sites (TSSs) in several species. Strikingly, ~72% of them could not be assigned to a specific gene and initiate at unconventional regions, outside promoters or enhancers. Here, we probe these unassigned TSSs and show that, in all species studied, a significant fraction of CAGE peaks initiate at microsatellites, also called short tandem repeats (STRs). To confirm this transcription, we develop Cap Trap RNA-seq, a technology which combines cap trapping and long read MinION sequencing. We train sequence-based deep learning models able to predict CAGE signal at STRs with high accuracy. These models unveil the importance of STR surrounding sequen2021-01-01T00:00:00Z2021 Jun2026-05-08T01:23:32.057Z2022-02-10T14:11:36.014ZS-EPMC81725403407888510.1038/s41467-021-23143-7