AI Insight
Researchers developed PatHoST, a computational framework that improves detection of pathogen locations in infected tissue using spatial transcriptomics data. The method uses machine learning to identify host gene expression patterns that predict where pathogens are located, overcoming the problem of incomplete pathogen detection in standard dual host-pathogen profiling. Testing on Mycobacterium abscessus-infected mouse lungs and human SARS-CoV-2 lung tissue demonstrated that PatHoST can accurately reconstruct pathogen spatial distributions and reveal localized host immune responses.
Why it matters
This approach enables better understanding of how infections develop within tissues by revealing pathogen locations and local host responses even when direct pathogen detection is limited. The method could improve studies of various infectious diseases and help identify targeted therapeutic strategies based on spatial infection patterns.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Dual host-pathogen spatial transcriptomics enables spatially resolved investigation of infection, yet the completeness of pathogen detection remains poorly characterized. Here, we generated pathogen-only spatial transcriptomic sections from a chronic Mycobacterium abscessus lung infection model to experimentally benchmark pathogen detection, revealing substantially increased pathogen transcript recovery compared with conventional dual profiling. Building on these data, we developed PatHoST, a weakly supervised computational framework that uses pathogen-derived information to generate silver-standard labels and learns host transcriptional programs predictive of pathogen presence. PatHoST reconstructed pathogen spatial distributions across independent tissue sections and enabled the identification of localized host regulatory programs beyond conventional cluster-based analyses. Application to human SARS-CoV-2 lung spatial transcriptomics supported generalization across host species and pathogen classes. Together, PatHoST combines purpose-built experimental benchmarking with weakly supervised inference to recover latent pathogen information and extend the analytical capabilities of dual host-pathogen spatial transcriptomics.
Source: PatHoST: weakly supervised reconstruction of pathogen spatial distributions from host transcriptomes