AI-guided discovery of the invariant host response to viral pandemics

Sahoo, D.; Katkar, G. D.; Khandelwal, S.; Behroozikhah, M.; Claire, A.; Castillo, V.; Tindle, C.; Fuller, M.; Taheri, S.; Rogers, T. F.; Beutler, N.; Ramirez, S.; Rawlings, S. A.; Pretorius, V.; Smith, D.; Burton, D. R.; Alexander, L. C.; Duran, J. M.; Crotty, S.; Dan, J. M.; Das, S.; Ghosh, P.

doi:10.1101/2020.09.21.305698

Home/Archive/bioRxiv

bioRxiv · 10.1101/2020.09.21.305698

AI-guided discovery of the invariant host response to viral pandemics

View original source ↗Export metadata ↓

Abstract

We sought to define the host immune response, a.k.a, the "cytokine storm" that has been implicated in fatal COVID-19 using an AI-based approach. Over 45,000 transcriptomic datasets of viral pandemics were analyzed to extract a 166-gene signature using ACE2 as a seed gene; ACE2 was rationalized because it encodes the receptor that facilitates the entry of SARS-CoV-2 (the virus that causes COVID-19) into host cells. Surprisingly, this 166-gene signature was conserved in all viral pandemics, including COVID-19, and a subset of 20-genes classified disease severity, inspiring the nomenclatures ViP and severe-ViP signatures, respectively. The ViP signatures pinpointed a paradoxical phenomenon wherein lung epithelial and myeloid cells mount an IL15 cytokine storm, and epithelial and NK cell senescence and apoptosis determines severity/fatality. Precise therapeutic goals were formulated and subsequently validated in high-dose SARS-CoV-2-challenged hamsters using neutralizing antibodies that abrogate SARS-CoV-2*ACE2 engagement or a directly acting antiviral agent, EIDD-2801. IL15/IL15RA were elevated in the lungs of patients with fatal disease, and plasma levels of the cytokine tracked with disease severity. Thus, the ViP signatures provide a quantitative and qualitative framework for titrating the immune response in viral pandemics and may serve as a powerful unbiased tool to rapidly assess disease severity and vet candidate drugs. One Sentence SummaryThe host immune response in COVID-19. PANEL: RESEARCH IN CONTEXTO_ST_ABSEvidence before this studyC_ST_ABSThe SARS-CoV-2 pandemic has inspired many groups to find innovative methodologies that can help us understand the host immune response to the virus; unchecked proportions of such immune response have been implicated in fatality. We searched GEO and ArrayExpress that provided many publicly available gene expression data that objectively measure the host immune response in diverse conditions. However, challenges remain in identifying a set of host response events that are common to every condition. There are no studies that provide a reproducible assessment of prognosticators of disease severity, the host response, and therapeutic goals. Consequently, therapeutic trials for COVID-19 have seen many more misses than hits. This work used multiple (> 45,000) gene expression datasets from GEO and ArrayExpress and analyzed them using an unbiased computational approach that relies upon fundamentals of gene expression patterns and mathematical precision when assessing them. Added value of this studyThis work identifies a signature that is surprisingly conserved in all viral pandemics, including Covid-19, inspiring the nomenclature ViP-signature. A subset of 20-genes classified disease severity in respiratory pandemics. The ViP signatures pinpointed the nature and source of the cytokine storm mounted by the host. They also helped formulate precise therapeutic goals and rationalized the repurposing of FDA-approved drugs. Implications of all the available evidenceThe ViP signatures provide a quantitative and qualitative framework for assessing the immune response in viral pandemics when creating pre-clinical models; they serve as a powerful unbiased tool to rapidly assess disease severity and vet candidate drugs.

Source connections

Published article ↗

Explore related subjects

bioinformatics Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Sahoo, D., Katkar, G. D., Khandelwal, S., Behroozikhah, M., Claire, A., Castillo, V., Tindle, C., Fuller, M., Taheri, S., Rogers, T. F., Beutler, N., Ramirez, S., Rawlings, S. A., Pretorius, V., Smith, D., Burton, D. R., Alexander, L. C., Duran, J. M., Crotty, S., Dan, J. M., Das, S., Ghosh, P.. 2020-09-22. AI-guided discovery of the invariant host response to viral pandemics. https://doi.org/10.1101/2020.09.21.305698

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

RoDPAA: A Robust Dissimilarity-Based Principal Angle Analysis Framework for Multi-View Microbiome Data Integration

Multiview microbiome studies profile the same participants across body sites or molecular layers, but integration is hampered by sparse, compositional data, modest sample sizes and methods that recognize only exact sharing. We present RoDPAA (Robust Dissimilarity-based Principal Angle Analysis), a framework that represents each view through any data-appropriate dissimilarity, such as Bray--Curtis, without log-ratio transformation or likelihood assumptions. RoDPAA aligns view-specific signal subspaces through principal angles, captures complete and partial sharing, and tests by permutation how many components are shared, using only eigendecomposition and little tuning. In Gaussian simulations, RoDPAA matched JIVE and MOFA given true ranks, ran faster and alone recovered the shared dimension in every replicate. In two inflammatory bowel disease (IBD) cohorts with sparse, zero-inflated data, Bray--Curtis-based RoDPAA recovered shared structure that was stable to prevalence filtering, whereas log-ratio-based alternatives were not. Its leading microbiome--metabolome factors, learned without disease labels, were the only ones associated with IBD.

bioinformatics↗

bioRxiv2026

Deducing splicing events from transcript sets with splicelogic

Long-read RNA sequencing captures a greater proportion of full-length transcripts, enabling more accurate isoform identification, discovery of novel isoforms, and more precise differential transcript usage analysis. However, comparing transcripts, whether between conditions or between sets such as novel and annotated isoforms, yields complex isoform changes that are hard to manipulate, summarize, and visualize. This is compounded by vertical integration of some tools, where downstream analysis depends on the specific output of a particular upstream tool. To address this gap, we have developed splicelogic, an R/Bioconductor package that makes transcript-set comparisons biologically interpretable by turning them into discrete splicing events, such as exon skipping, alternative splice sites, intron retention, and alternative transcription start and end sites. We designed splicelogic to integrate easily with existing RNA-seq workflows and tools. The splicelogic functions are modular, readable, and extensible, making it easy for users to adapt core logic to their own custom scientific questions or novel splicing definitions. By obtaining interpretable splicing events from transcript changes, splicelogic enables downstream analyses, including annotation of RNA-binding protein recognition sites, alongside summary statistics and visualization. The splicelogic package is publicly available and distributed through Bioconductor.

bioinformatics↗

bioRxiv2026

Contrastive Learning Improves Generalization on M2OR Olfactory Binding Dataset

Olfaction-based biosensors stand to have profound impacts in fields such as medical diagnostics, national security, and drug discovery. A promising mode of vapor phase biodetection mimics volatile organic compound-olfactory receptor (VOC-OR) interactions with target-specific protein or peptide probes; however, lack of experimental coverage remains a core bottleneck. Current computational approaches aim to improve prediction of VOC-OR binding to reduce experimental burden, and current machine learning approaches fail to demonstrate a level of generalization that addresses current engineering challenges. Here, using the Molecular to Olfactory Dataset (M2OR) and Insect Olfactory Binding Protein Database (iOBPdb) datasets, data partitioning methods and their performance are evaluated on a neural network and contrastive learning framework on such VOC-OR binding datasets, varying: model training, architecture, and structure of the input data distribution. It was found that holding out certain chemical or proteomic motifs from the training data diminished model performance with a two layer neural network model. The model was able to generalize to state of the art performance (0.88 AUCROC) on every data split by applying a contrastive learning objective defined by shared molecular constituents and a stochastic sampler. This suggests the contrastive objective provides a strong inductive bias for neural networks to predict novel VOC-OR pairs. These findings highlight that model performance on pairwise biological interaction datasets can be inflated through identity leakage from constituent overlap, and emphasize the need for more robust evaluation protocols. Furthermore, we show contrastive regularization mitigates dataset degeneracy through enforcement of compatibility in the models latent geometry that may extend to other binder target-datasets where identity leakage confounds evaluation.

bioinformatics↗