A transfer-learning approach to predict antigen immunogenicity and T-cell receptor specificity
Abstract
Antigen immunogenicity and the specificity of binding of T-cell receptors to antigens are key properties underlying effective immune responses. Here we propose diffRBM, an approach based on transfer learning and Restricted Boltzmann Machines, to build sequence-based predictive models of these properties. DiffRBM is designed to learn the distinctive patterns in amino-acid composition that, on the one hand, underlie the antigen's probability of triggering a response, and on the other hand the T-cell receptor's ability to bind to a given antigen. We show that the patterns learnt by diffRBM allow us to predict putative contact sites of the antigen-receptor complex. We also discriminate immunogenic and non-immunogenic antigens, antigen-specific and generic receptors, reaching performances that compare favorably to existing sequence-based predictors of antigen immunogenicity and T-cell receptor specificity.
Data availability
The current manuscript is a computational study, so no data have been generated for this manuscript. The data used are downloaded from public databases. The pre-processed data, the results of the analysis, the codes to train and evaluate the models as well as the trained models are all available at the github page https://github.com/bravib/diffRBM_immunogenicity_TCRspecificity.
Article and author information
Author details
Funding
Agence Nationale de la Recherche (RBMPro CE30-0021-01)
- Andrea Di Gioacchino
- Jorge Fernandez-de-Cossio-Diaz
European Research Council (COG 724208)
- Jorge Fernandez-de-Cossio-Diaz
- Aleksandra M Walczak
HORIZON EUROPE Marie Sklodowska-Curie Actions (101026293)
- Andrea Di Gioacchino
Agence Nationale de la Recherche (RBMPro CE30-0021-01)
- Simona Cocco
- Rémi Monasson
Agence Nationale de la Recherche (Prodigen)
- Andrea Di Gioacchino
- Jorge Fernandez-de-Cossio-Diaz
Agence Nationale de la Recherche (Prodigen)
- Simona Cocco
- Rémi Monasson
Agence Nationale de la Recherche (Decrypted CE30-0021-01)
- Andrea Di Gioacchino
- Jorge Fernandez-de-Cossio-Diaz
Agence Nationale de la Recherche (Decrypted CE30-0021-01)
- Simona Cocco
- Rémi Monasson
Agence Nationale de la Recherche (RESP-REP CE45-0018)
- Aleksandra M Walczak
- Thierry Mora
Agence Nationale de la Recherche (RESP-REP CE45-0018)
- Barbara Bravi
European Research Council (COG 724208)
- Barbara Bravi
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Reviewing Editor
- Anne-Florence Bitbol, Ecole Polytechnique Federale de Lausanne (EPFL), Switzerland
Version history
- Received: November 23, 2022
- Accepted: September 7, 2023
- Accepted Manuscript published: September 8, 2023 (version 1)
Copyright
© 2023, Bravi et al.
This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 175
- Page views
-
- 71
- Downloads
-
- 0
- Citations
Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.
Download links
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Further reading
-
- Computational and Systems Biology
- Neuroscience
Cerebellar climbing fibers convey diverse signals, but how they are organized in the compartmental structure of the cerebellar cortex during learning remains largely unclear. We analyzed a large amount of coordinate-localized two-photon imaging data from cerebellar Crus II in mice undergoing 'Go/No-go' reinforcement learning. Tensor component analysis revealed that a majority of climbing fiber inputs to Purkinje cells were reduced to only four functional components, corresponding to accurate timing control of motor initiation related to a Go cue, cognitive error-based learning, reward processing, and inhibition of erroneous behaviors after a No-go cue. Changes in neural activities during learning of the first two components were correlated with corresponding changes in timing control and error learning across animals, indirectly suggesting causal relationships. Spatial distribution of these components coincided well with boundaries of Aldolase-C/zebrin II expression in Purkinje cells, whereas several components are mixed in single neurons. Synchronization within individual components was bidirectionally regulated according to specific task contexts and learning stages. These findings suggest that, in close collaborations with other brain regions including the inferior olive nucleus, the cerebellum, based on anatomical compartments, reduces dimensions of the learning space by dynamically organizing multiple functional components, a feature that may inspire new-generation AI designs.
-
- Computational and Systems Biology
- Genetics and Genomics
Tissue fibrosis affects multiple organs and involves a master-regulatory role of macrophages which respond to an initial inflammatory insult common in all forms of fibrosis. The recently unravelled multi-organ heterogeneity of macrophages in healthy and fibrotic human disease suggests that macrophages expressing osteopontin (SPP1), associate with lung and liver fibrosis. However, the conservation of this SPP1+ macrophage population across different tissues, and its specificity to fibrotic diseases with different etiologies remain unclear. Integrating 15 single cell RNA-sequencing datasets to profile 235,930 tissue macrophages from healthy and fibrotic heart, lung, liver, kidney, skin and endometrium, we extended the association of SPP1+ macrophages with fibrosis to all these tissues. We also identified a subpopulation expressing matrisome-associated genes (e.g., matrix metalloproteinases and their tissue inhibitors), functionally enriched for ECM remodelling and cell metabolism, representative of a matrisome-associated macrophage (MAM) polarization state within SPP1+ macrophages. Importantly, the MAM polarization state follows a differentiation trajectory from SPP1+ macrophages and is associated with a core set of regulon activity. SPP1+ macrophages without the MAM polarization state (SPP1+MAM-) show a positive association with ageing lung in mice and humans. These results suggest an advanced and conserved polarization state of SPP1+ macrophages in fibrotic tissues resulting from prolonged inflammatory cues within each tissue microenvironment.