Mining adaptive immune receptor repertoires for biological and clinical information using machine learning

Victor Greiff; Gur Yaari; Lindsay G. Cowell

doi:10.1016/j.coisb.2020.10.010

Mining adaptive immune receptor repertoires for biological and clinical information using machine learning

Victor Greiff, Gur Yaari, Lindsay G. Cowell

Bar-Ilan University - The Alexander Kofkin Faculty of Engineering

Research output: Contribution to journal › Review article › peer-review

65 Scopus citations

Abstract

The adaptive immune system stores invaluable information about current and past immune responses and may serve as an ultrasensitive biosensor. Given the immune system's critical role in a wide variety of disease types, this has broad implications for biomedicine. Machine and deep learning is being leveraged to decipher how information is encoded in adaptive immune receptor repertoires to enable prediction from adaptive immune responses and fast-track vaccine, therapeutics, and diagnostics development. Recent advances include predicting the presence of immunity after vaccination or infection, predicting the presence of disease, and designing antibody-based therapeutics. Outstanding challenges encompass increasing our knowledge of the feature space structure that encodes relevant immune information, addressing the lack of ground truth–labeled data, and improving our handling of genetic and environmental confounding factors.

Original language	English
Pages (from-to)	109-119
Number of pages	11
Journal	Current Opinion in Systems Biology
Volume	24
DOIs	https://doi.org/10.1016/j.coisb.2020.10.010
State	Published - Dec 2020

Bibliographical note

Publisher Copyright:
© 2020 The Authors

Funding

V.G., L.G.C., and G.Y. are supported by European Union's Horizon 2020 Research and Innovation Program as members of the iReceptor Plus Consortium (#825821). The contents of this document are the sole responsibility of the authors and can under no circumstances be regarded as reflecting the position of the European Union. V.G. is also supported by the UiO World-Leading Research Community , the UiO: Life Sciences Convergence Environment Immunolingo , The Helmsley Charitable Trust ( #2019PG-T1D011 ), and a Research Council of Norway FRIPRO project ( #300740 ). G.Y. is also supported by the Israel Science Foundation (ISF [ 832/16 ]). L.G.C. is also supported by the Simmons Comprehensive Cancer Center , Be the Difference Foundation , Young Texans Against Cancer , and Commercial Real Estate Women of Dallas (CREW Dallas).

Funders	Funder number
Difference Foundation
Simmons Comprehensive Cancer Center
UiO World-Leading Research Community
Young Texans Against Cancer
Leona M. and Harry B. Helmsley Charitable Trust	2019PG-T1D011
Horizon 2020 Framework Programme
European Commission	825821
Israel Science Foundation	832/16
Norges Forskningsråd	300740

Keywords

AIRR-seq
Adaptive immune receptor repertoires
Immunodiagnostics
Machine learning

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.coisb.2020.10.010

Cite this

@article{dd15df6fb53f40e9ba6167808599a13d,

title = "Mining adaptive immune receptor repertoires for biological and clinical information using machine learning",

abstract = "The adaptive immune system stores invaluable information about current and past immune responses and may serve as an ultrasensitive biosensor. Given the immune system's critical role in a wide variety of disease types, this has broad implications for biomedicine. Machine and deep learning is being leveraged to decipher how information is encoded in adaptive immune receptor repertoires to enable prediction from adaptive immune responses and fast-track vaccine, therapeutics, and diagnostics development. Recent advances include predicting the presence of immunity after vaccination or infection, predicting the presence of disease, and designing antibody-based therapeutics. Outstanding challenges encompass increasing our knowledge of the feature space structure that encodes relevant immune information, addressing the lack of ground truth–labeled data, and improving our handling of genetic and environmental confounding factors.",

keywords = "AIRR-seq, Adaptive immune receptor repertoires, Immunodiagnostics, Machine learning",

author = "Victor Greiff and Gur Yaari and Cowell, \{Lindsay G.\}",

note = "Publisher Copyright: {\textcopyright} 2020 The Authors",

year = "2020",

month = dec,

doi = "10.1016/j.coisb.2020.10.010",

language = "אנגלית",

volume = "24",

pages = "109--119",

journal = "Current Opinion in Systems Biology",

issn = "2452-3100",

publisher = "Elsevier Ltd.",

}

TY - JOUR

T1 - Mining adaptive immune receptor repertoires for biological and clinical information using machine learning

AU - Greiff, Victor

AU - Yaari, Gur

AU - Cowell, Lindsay G.

PY - 2020/12

Y1 - 2020/12

N2 - The adaptive immune system stores invaluable information about current and past immune responses and may serve as an ultrasensitive biosensor. Given the immune system's critical role in a wide variety of disease types, this has broad implications for biomedicine. Machine and deep learning is being leveraged to decipher how information is encoded in adaptive immune receptor repertoires to enable prediction from adaptive immune responses and fast-track vaccine, therapeutics, and diagnostics development. Recent advances include predicting the presence of immunity after vaccination or infection, predicting the presence of disease, and designing antibody-based therapeutics. Outstanding challenges encompass increasing our knowledge of the feature space structure that encodes relevant immune information, addressing the lack of ground truth–labeled data, and improving our handling of genetic and environmental confounding factors.

AB - The adaptive immune system stores invaluable information about current and past immune responses and may serve as an ultrasensitive biosensor. Given the immune system's critical role in a wide variety of disease types, this has broad implications for biomedicine. Machine and deep learning is being leveraged to decipher how information is encoded in adaptive immune receptor repertoires to enable prediction from adaptive immune responses and fast-track vaccine, therapeutics, and diagnostics development. Recent advances include predicting the presence of immunity after vaccination or infection, predicting the presence of disease, and designing antibody-based therapeutics. Outstanding challenges encompass increasing our knowledge of the feature space structure that encodes relevant immune information, addressing the lack of ground truth–labeled data, and improving our handling of genetic and environmental confounding factors.

KW - AIRR-seq

KW - Adaptive immune receptor repertoires

KW - Immunodiagnostics

KW - Machine learning

UR - https://www.scopus.com/pages/publications/85096706165

U2 - 10.1016/j.coisb.2020.10.010

DO - 10.1016/j.coisb.2020.10.010

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.systematicreview???

AN - SCOPUS:85096706165

SN - 2452-3100

VL - 24

SP - 109

EP - 119

JO - Current Opinion in Systems Biology

JF - Current Opinion in Systems Biology

ER -

Mining adaptive immune receptor repertoires for biological and clinical information using machine learning

Abstract

Bibliographical note

Funding

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this