Nature Communications (Aug 2024)
Community assessment of methods to deconvolve cellular composition from bulk gene expression
- Brian S. White,
- Aurélien de Reyniès,
- Aaron M. Newman,
- Joshua J. Waterfall,
- Andrew Lamb,
- Florent Petitprez,
- Yating Lin,
- Rongshan Yu,
- Martin E. Guerrero-Gimenez,
- Sergii Domanskyi,
- Gianni Monaco,
- Verena Chung,
- Jineta Banerjee,
- Daniel Derrick,
- Alberto Valdeolivas,
- Haojun Li,
- Xu Xiao,
- Shun Wang,
- Frank Zheng,
- Wenxian Yang,
- Carlos A. Catania,
- Benjamin J. Lang,
- Thomas J. Bertus,
- Carlo Piermarocchi,
- Francesca P. Caruso,
- Michele Ceccarelli,
- Thomas Yu,
- Xindi Guo,
- Julie Bletz,
- John Coller,
- Holden Maecker,
- Caroline Duault,
- Vida Shokoohi,
- Shailja Patel,
- Joanna E. Liliental,
- Stockard Simon,
- Tumor Deconvolution DREAM Challenge consortium,
- Julio Saez-Rodriguez,
- Laura M. Heiser,
- Justin Guinney,
- Andrew J. Gentles
Affiliations
- Brian S. White
- Sage Bionetworks
- Aurélien de Reyniès
- Centre de Recherche des Cordeliers, INSERM U1138, Université Paris Cité
- Aaron M. Newman
- Institute for Stem Cell Biology and Regenerative Medicine, Stanford University
- Joshua J. Waterfall
- INSERM U830 and Translational Research Department, Institut Curie, PSL Research University
- Andrew Lamb
- Sage Bionetworks
- Florent Petitprez
- Programme Cartes d’Identité des Tumeurs, Ligue Nationale Contre le Cancer
- Yating Lin
- Xiamen University
- Rongshan Yu
- Xiamen University
- Martin E. Guerrero-Gimenez
- Institute of Biochemistry and Biotechnology, School of Medicine, National University of Cuyo
- Sergii Domanskyi
- Michigan State University
- Gianni Monaco
- BIOGEM Institute of Molecular Biology and Genetics
- Verena Chung
- Sage Bionetworks
- Jineta Banerjee
- Sage Bionetworks
- Daniel Derrick
- Department of Biomedical Engineering, Knight Cancer Institute, Oregon Health & Science University
- Alberto Valdeolivas
- Heidelberg University, Faculty of Medicine, and Heidelberg University Hospital, Institute for Computational Biomedicine
- Haojun Li
- Xiamen University
- Xu Xiao
- Xiamen University
- Shun Wang
- Department of Pathology, Cancer Hospital, Chinese Aacdemy of Medical Science
- Frank Zheng
- AmoyDx
- Wenxian Yang
- Aginome Scientific
- Carlos A. Catania
- Laboratory of Intelligent Systems (LABSIN), Engineering School, National University of Cuyo
- Benjamin J. Lang
- Department of Radiation Oncology, Beth Israel Deaconess Medical Center, Harvard Medical School
- Thomas J. Bertus
- Michigan State University
- Carlo Piermarocchi
- Michigan State University
- Francesca P. Caruso
- BIOGEM Institute of Molecular Biology and Genetics
- Michele Ceccarelli
- BIOGEM Institute of Molecular Biology and Genetics
- Thomas Yu
- Sage Bionetworks
- Xindi Guo
- Sage Bionetworks
- Julie Bletz
- Sage Bionetworks
- John Coller
- Stanford Functional Genomics Facility, Stanford University School of Medicine
- Holden Maecker
- Institute for Immunity, Transplantation, and Infection, Stanford University School of Medicine
- Caroline Duault
- Institute for Immunity, Transplantation, and Infection, Stanford University School of Medicine
- Vida Shokoohi
- Stanford Functional Genomics Facility, Stanford University School of Medicine
- Shailja Patel
- Translational Applications Service Center, Stanford University School of Medicine
- Joanna E. Liliental
- Translational Applications Service Center, Stanford University School of Medicine
- Stockard Simon
- Sage Bionetworks
- Tumor Deconvolution DREAM Challenge consortium
- Julio Saez-Rodriguez
- Heidelberg University, Faculty of Medicine, and Heidelberg University Hospital, Institute for Computational Biomedicine
- Laura M. Heiser
- Department of Biomedical Engineering, Knight Cancer Institute, Oregon Health & Science University
- Justin Guinney
- Sage Bionetworks
- Andrew J. Gentles
- Department of Biomedical Data Science, Stanford University
- DOI
- https://doi.org/10.1038/s41467-024-50618-0
- Journal volume & issue
-
Vol. 15,
no. 1
pp. 1 – 22
Abstract
Abstract We evaluate deconvolution methods, which infer levels of immune infiltration from bulk expression of tumor samples, through a community-wide DREAM Challenge. We assess six published and 22 community-contributed methods using in vitro and in silico transcriptional profiles of admixed cancer and healthy immune cells. Several published methods predict most cell types well, though they either were not trained to evaluate all functional CD8+ T cell states or do so with low accuracy. Several community-contributed methods address this gap, including a deep learning-based approach, whose strong performance establishes the applicability of this paradigm to deconvolution. Despite being developed largely using immune cells from healthy tissues, deconvolution methods predict levels of tumor-derived immune cells well. Our admixed and purified transcriptional profiles will be a valuable resource for developing deconvolution methods, including in response to common challenges we observe across methods, such as sensitive identification of functional CD4+ T cell states.
