RT Journal Article
SR Electronic
T1 immuneML: an ecosystem for machine learning analysis of adaptive immune receptor repertoires
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 2021.03.08.433891
DO 10.1101/2021.03.08.433891
A1 Milena Pavlović
A1 Lonneke Scheffer
A1 Keshav Motwani
A1 Chakravarthi Kanduri
A1 Radmila Kompova
A1 Nikolay Vazov
A1 Knut Waagan
A1 Fabian L. M. Bernal
A1 Alexandre Almeida Costa
A1 Brian Corrie
A1 Rahmad Akbar
A1 Ghadi S. Al Hajj
A1 Gabriel Balaban
A1 Todd M. Brusko
A1 Maria Chernigovskaya
A1 Scott Christley
A1 Lindsay G. Cowell
A1 Robert Frank
A1 Ivar Grytten
A1 Sveinung Gundersen
A1 Ingrid Hobæk Haff
A1 Sepp Hochreiter
A1 Eivind Hovig
A1 Ping-Han Hsieh
A1 Günter Klambauer
A1 Marieke L. Kuijjer
A1 Christin Lund-Andersen
A1 Antonio Martini
A1 Thomas Minotto
A1 Johan Pensar
A1 Knut Rand
A1 Enrico Riccardi
A1 Philippe A. Robert
A1 Artur Rocha
A1 Andrei Slabodkin
A1 Igor Snapkov
A1 Ludvig M. Sollid
A1 Dmytro Titov
A1 Cédric R. Weber
A1 Michael Widrich
A1 Gur Yaari
A1 Victor Greiff
A1 Geir Kjetil Sandve
YR 2021
UL http://biorxiv.org/content/early/2021/03/15/2021.03.08.433891.abstract
AB Adaptive immune receptor repertoires (AIRR) are key targets for biomedical research as they record past and ongoing adaptive immune responses. The capacity of machine learning (ML) to identify complex discriminative sequence patterns renders it an ideal approach for AIRR-based diagnostic and therapeutic discovery. To date, widespread adoption of AIRR ML has been inhibited by a lack of reproducibility, transparency, and interoperability. immuneML (immuneml.uio.no) addresses these concerns by implementing each step of the AIRR ML process in an extensible, open-source software ecosystem that is based on fully specified and shareable workflows. To facilitate widespread user adoption, immuneML is available as a command-line tool and through an intuitive Galaxy web interface, and extensive documentation of workflows is provided. We demonstrate the broad applicability of immuneML by (i) reproducing a large-scale study on immune state prediction, (ii) developing, integrating, and applying a novel method for antigen specificity prediction, and (iii) showcasing streamlined interpretability-focused benchmarking of AIRR ML.Competing Interest StatementVG declares advisory board positions in aiNET GmbH and Enpicom B.V.