Discovery of widespread transcription initiation at microsatellites predictable by sequence-based deep neural network

FANTOM consortium

doi:10.1038/s41467-021-23143-7

Discovery of widespread transcription initiation at microsatellites predictable by sequence-based deep neural network

FANTOM consortium

Institut de Biologie Computationnelle
Institute of Molecular Genetics of Montpellier
Sanofi-Aventis
RIKEN
National Institute of Advanced Industrial Science and Technology
The University of Tokyo
Waseda University
University of Edinburgh
European Molecular Biology Laboratory
Wellcome Trust Sanger Institute
King Abdullah University of Science and Technology
McGill University
University of New South Wales
University of Western Australia
Columbia University
University of Copenhagen
Juntendo University
University of California at Berkeley
Karolinska Institutet
Charité – Universitätsmedizin Berlin
Jackson Laboratory
Agency for Science, Technology and Research, Singapore
Stanford University
University of Queensland
University of Udine
National University of Singapore
Oregon State University
Istituto Nazionale di Genetica Molecolare
Research Organization of Information and Systems
University of Basel
Swiss Institute of Bioinformatics
International Centre for Genetic Engineering and Biotechnology
University of Cape Town
Lawrence Berkeley National Laboratory
National Institutes of Health
Max Delbrück Center for Molecular Medicine in the Helmholtz Association
Technische Universität Dresden
Sorbonne Université
Telethon Kids Institute
University of British Columbia
Fondazione Bruno Kessler
The Children's Hospital at Westmead
Laboratorio Nazionale Consorzio Italiano Biotecnologie (LNCIB)
Cold Spring Harbor Laboratory
Vrije Universiteit Amsterdam
Massey University
Swiss Federal Institute of Technology Lausanne
ETH Zurich—Institute for Particle Physics and Astrophysics (IPA)
Kazan Volga Region Federal University
Fondazione Telethon
SUNY Buffalo
Harvard University
Centre for Genomic Regulation
National Center for Global Health and Medicine
Norwegian University of Science and Technology
University of Regensburg
Regensburg Centre for Interventional Immunology
Stockholm University
Kobe University
University of Miami
University of Delaware
Rice University
Russian Academy of Sciences
Johns Hopkins University
Imperial College London
KTH Royal Institute of Technology
The University of Osaka
University of Amsterdam
The Systems Biology Institute
Technische UniversitŁt Mnchen
University of Bristol
University of Helsinki
International School for Advanced Studies
Italian Institute of Technology
University of Bergen
Wistar Institute
Leiden University
German Center for Neurodegenerative Diseases
University of Sheffield
University of Melbourne
Walter and Eliza Hall Institute of Medical Research
Kyoto University
Nagahama Institute of Bio-Science and Technology
Research Organization of Information and Systems, National Institute of Genetics Mishima
Huaqiao University
St. Laurent Institute
Lomonosov Moscow State University
Kyoto Prefectural University of Medicine
Folkhalsan
Sony Group Corporation
Monash University
Okinawa Institute of Science and Technology Graduate University
University of Nottingham
Keio University
RAS - Engelhardt Institute of Molecular Biology
Skolkovo Institute of Science and Technology
Tohoku University
Kyungpook National University
Wayne State University
Moscow Institute of Physics and Technology
Albert Einstein College of Medicine
IMPPC
Kanagawa Cancer Center Research Institute
Saitama Medical University
Yokohama City University
Ohu University
Lund University
Biomedical Primate Research Centre
Uppsala University
Christiana Care Health System
Dana-Farber Cancer Institute
Duke-NUS Medical School
Guy's and St Thomas' NHS Foundation Trust
Technical University of Munich
Paracelsus Medical University
Institute of Science Tokyo
Kumamoto University
Icahn School of Medicine at Mount Sinai
Brown University
Kyushu University
Tokyo University of Pharmacy and Life Sciences
Netherlands Institute for Developmental Biology
Nihon University
Osaka Institute of Technology
Memorial Sloan-Kettering Cancer Center
University of Eastern Piedmont

Research output: Contribution to journal › Article › peer-review

20 Citations (Scopus)

Abstract

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 sequences not only to distinguish STR classes, but also to predict the level of transcription initiation. Importantly, genetic variants linked to human diseases are preferentially found at STRs with high transcription initiation level, supporting the biological and clinical relevance of transcription initiation at STRs. Together, our results extend the repertoire of non-coding transcription associated with DNA tandem repeats and complexify STR polymorphism.

Original language	English
Article number	3297
Journal	Nature Communications
Volume	12
Issue number	1
DOIs	https://doi.org/10.1038/s41467-021-23143-7
Publication status	Published - 1 Dec 2021
Externally published	Yes

Access to Document

10.1038/s41467-021-23143-7

Cite this

@article{28d05af7013e4e31bc083b71e16985e7,

title = "Discovery of widespread transcription initiation at microsatellites predictable by sequence-based deep neural network",

abstract = "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, \textasciitilde{}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 sequences not only to distinguish STR classes, but also to predict the level of transcription initiation. Importantly, genetic variants linked to human diseases are preferentially found at STRs with high transcription initiation level, supporting the biological and clinical relevance of transcription initiation at STRs. Together, our results extend the repertoire of non-coding transcription associated with DNA tandem repeats and complexify STR polymorphism.",

author = "\{FANTOM consortium\} and Mathys Grapotte and Manu Saraswat and Chlo{\'e} Bessi{\`e}re and Christophe Menichelli and Ramilowski, \{Jordan A.\} and Jessica Severin and Yoshihide Hayashizaki and Masayoshi Itoh and Michihira Tagami and Mitsuyoshi Murata and Miki Kojima-Ishiyama and Shohei Noma and Shuhei Noguchi and Takeya Kasukawa and Akira Hasegawa and Harukazu Suzuki and Hiromi Nishiyori-Sueki and Frith, \{Martin C.\} and Imad Abugessaisa and Stuart Aitken and Aken, \{Bronwen L.\} and Intikhab Alam and Tanvir Alam and Rami Alasiri and Alhendi, \{Ahmad M.N.\} and Hamid Alinejad-Rokny and Alvarez, \{Mariano J.\} and Robin Andersson and Takahiro Arakawa and Marito Araki and Taly Arbel and John Archer and Archibald, \{Alan L.\} and Erik Arner and Peter Arner and Kiyoshi Asai and Haitham Ashoor and Gaby Astrom and Magda Babina and Baillie, \{J. Kenneth\} and Bajic, \{Vladimir B.\} and Archana Bajpai and Sarah Baker and Baldarelli, \{Richard M.\} and Adam Balic and Mukesh Bansal and Batagov, \{Arsen O.\} and Serafim Batzoglou and Beckhouse, \{Anthony G.\} and Beltrami, \{Antonio P.\}",

note = "Publisher Copyright: {\textcopyright} 2021, The Author(s).",

year = "2021",

month = dec,

day = "1",

doi = "10.1038/s41467-021-23143-7",

language = "English",

volume = "12",

journal = "Nature Communications",

issn = "2041-1723",

publisher = "Nature Research",

number = "1",

}

TY - JOUR

T1 - Discovery of widespread transcription initiation at microsatellites predictable by sequence-based deep neural network

AU - FANTOM consortium

AU - Grapotte, Mathys

AU - Saraswat, Manu

AU - Bessière, Chloé

AU - Menichelli, Christophe

AU - Ramilowski, Jordan A.

AU - Severin, Jessica

AU - Hayashizaki, Yoshihide

AU - Itoh, Masayoshi

AU - Tagami, Michihira

AU - Murata, Mitsuyoshi

AU - Kojima-Ishiyama, Miki

AU - Noma, Shohei

AU - Noguchi, Shuhei

AU - Kasukawa, Takeya

AU - Hasegawa, Akira

AU - Suzuki, Harukazu

AU - Nishiyori-Sueki, Hiromi

AU - Frith, Martin C.

AU - Abugessaisa, Imad

AU - Aitken, Stuart

AU - Aken, Bronwen L.

AU - Alam, Intikhab

AU - Alam, Tanvir

AU - Alasiri, Rami

AU - Alhendi, Ahmad M.N.

AU - Alinejad-Rokny, Hamid

AU - Alvarez, Mariano J.

AU - Andersson, Robin

AU - Arakawa, Takahiro

AU - Araki, Marito

AU - Arbel, Taly

AU - Archer, John

AU - Archibald, Alan L.

AU - Arner, Erik

AU - Arner, Peter

AU - Asai, Kiyoshi

AU - Ashoor, Haitham

AU - Astrom, Gaby

AU - Babina, Magda

AU - Baillie, J. Kenneth

AU - Bajic, Vladimir B.

AU - Bajpai, Archana

AU - Baker, Sarah

AU - Baldarelli, Richard M.

AU - Balic, Adam

AU - Bansal, Mukesh

AU - Batagov, Arsen O.

AU - Batzoglou, Serafim

AU - Beckhouse, Anthony G.

AU - Beltrami, Antonio P.

PY - 2021/12/1

Y1 - 2021/12/1

N2 - 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 sequences not only to distinguish STR classes, but also to predict the level of transcription initiation. Importantly, genetic variants linked to human diseases are preferentially found at STRs with high transcription initiation level, supporting the biological and clinical relevance of transcription initiation at STRs. Together, our results extend the repertoire of non-coding transcription associated with DNA tandem repeats and complexify STR polymorphism.

AB - 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 sequences not only to distinguish STR classes, but also to predict the level of transcription initiation. Importantly, genetic variants linked to human diseases are preferentially found at STRs with high transcription initiation level, supporting the biological and clinical relevance of transcription initiation at STRs. Together, our results extend the repertoire of non-coding transcription associated with DNA tandem repeats and complexify STR polymorphism.

UR - https://www.scopus.com/pages/publications/85107388625

U2 - 10.1038/s41467-021-23143-7

DO - 10.1038/s41467-021-23143-7

M3 - Article

C2 - 34078885

AN - SCOPUS:85107388625

SN - 2041-1723

VL - 12

JO - Nature Communications

JF - Nature Communications

IS - 1

M1 - 3297

ER -

Discovery of widespread transcription initiation at microsatellites predictable by sequence-based deep neural network

Abstract

Access to Document

Other files and links

Fingerprint

Cite this