Turner, A. P. | Search bioRxiv

bioRxiv2024

Single-cell Masked Autoencoder: An Accurate and Interpretable Automated Immunophenotyper

High-throughput single-cell cytometry data are crucial for understanding involvement of immune system in diseases and responses to treatment. Traditional methods for annotating cytometry data, specifically manual gating and clustering, face challenges in scalability, robustness, and accuracy. In this study, we propose a cytometry masked autoencoder (cyMAE), which offers an automated solution for immunophenotyping tasks including cell type annotation. The cyMAE model is designed to uphold user-defined cell type definitions, thereby facilitating easier interpretation and cross-study comparisons. The cyMAE model operates on a pre-train and fine-tune approach. In the pre-training phase, cyMAE employs Masked Cytometry Modelling (MCM) to learn relationships between protein markers in immune cells solely based on protein expression, without relying on prior information such as cell identity and cell type-specific marker proteins. Subsequently, the pre-trained cyMAE is fine-tuned on multiple specialized tasks via task-specific supervised learning. The pre-trained cyMAE addresses the shortcomings of manual gating and clustering methods by providing accurate and interpretable predictions. Through validation across multiple cohorts, we demonstrate that cyMAE effectively identifies co-occurrence patterns of bound labeled antibodies, delivers accurate and interpretable cellular immunophenotyping, and improves the prediction of subject metadata status. Specifically, we evaluated cyMAE for cell type annotation and imputation at the cellular-level and SARS-CoV-2 infection prediction, secondary immune response prediction against COVID-19, and prediction of the infection stage in COVID-19 progression at the subject-level. The introduction of cyMAE marks a significant step forward in immunology research, particularly in large-scale and high-throughput human immune profiling. This approach offers new possibilities for predicting and interpreting cellular-level and subject-level phenotypes in both health and disease.

bioinformatics↗

bioRxiv2019

Abiotic environmental adaptation in vertebrates is characterized by functional genomic constraint

Understanding the genomic basis of adaptation to different abiotic environments is important for understanding organismal responses to current short-term environmental fluctuations. Using functional and comparative genomics approaches, we here investigated whether genomic adaptation to a set of environmental parameters is contingent across vertebrate genomes or, alternatively, contains an element of evolutionary constraint that would be evident through recurrent involvement of specific subsets of genes and functions in adaptation to similar environments. We first identified 200 genes with signatures of selection from transcriptomes of 24 species of lacertid lizards with known adaptations in preferred temperature, correlated with thermal environment experienced by these lizards in their range. In order to discern genes adapting to climate from other selective factors, we then performed a meta-analysis of 1100 genes with signatures of selection obtained from-omics studies in vertebrate species adapted to different abiotic environments. We found that this gene set formed a tightly connected interactome which was to 23% enriched in predicted functions of adaptation to climate and to 18% involved in organismal stress response. We found a much higher degree of recurrent use of identical genes (43.6%) and functional similarity than expected by chance, and no clear division between genes used in ectotherm and endotherm physiological strategies. 171 out of 200 genes of Lacertidae were part of this network, indicating that a comparative genomic approach can help to disentangle genes functionally related to adaptation to different abiotic environments from other selective factors. These results furthermore highlight an important role of genomic constraint in adaptation to the abiotic environment, and narrows the set of candidate markers to be used in future research on environmental adaptability related to climate change.\n\nSignificance Statement / Nontechnical summaryWhile the convergent evolution of phenotypes in similar environments is a well-studied phenomenon, the genomic basis of such common phenotypes and physiologies is still enigmatic. The prevalent notion is that re-use of the same genes to adapt to similar environments in different species is about as likely as winning the lottery - but organismal systems are also, to some extent, comparable between different species such as man and fruit fly through shared genes and gene functions. In this paper, we test whether constraint or contingency is more prevalent in genomic adaptation of vertebrates to aspects of their abiotic environment. We find evidence for strong functional constraint and stress responsiveness of the genes involved, which might help understand how currently experienced stress under changing climates may result in future adaptation.

evolutionary biology↗