Biology subjects

Pancheva, A.

Publications and source records attributed to Pancheva, A..

3 recordsLinked to original sources

bioRxiv2025

Self-supervised AI reveals a lethal discohesive phenotype in lung adenocarcinoma

Applications of artificial intelligence (AI) to histopathology are now common, but most require supervision which inherently limits their scope. By using self-supervised learning (SSL), we discover and quantify the full range of histopathological appearances in a disease, and associate them with clinicopathological ground truths such as prognosis. We used this approach to discover under-appreciated morphologies of lung adenocarcinoma (LUAD), using a highly characterised resected tumour cohort of over 4000 slides from over 1000 patients. By constructing an authoritative lexicon of recurrent LUAD appearances, we ab initio discovered several stromal morphologies strongly predictive of outcome. With multimodal data integration and external dataset validation, we propose that epithelial discohesion is lethal, but only in the context of immunologically cold stroma. Both these morphological features are independent of current prognostic schema. Crucially, we describe these features in the context of real-world diagnostic histopathology, giving them immediate clinical translatability. Statement of SignificanceHistopathology relies heavily on epithelial morphology, often neglecting the stroma. We use self-supervised AI to identify under-appreciated morphologies in LUAD linked to poor outcome and validate these observations in an external cohort, demonstrating the utility of self-supervised AI as a powerful biological discovery tool.

cancer biology↗

bioRxiv2024

PLM-interact: extending protein language models to predict protein-protein interactions

Computational prediction of protein structure from amino acid sequences alone has been achieved with unprecedented accuracy, yet the prediction of protein-protein interactions (PPIs) remains an outstanding challenge. Here we assess the ability of protein language models (PLMs), routinely applied to protein folding, to be retrained for PPI prediction. Existing PPI prediction models that exploit PLMs use a pre-trained PLM feature set, ignoring that the proteins are physically interacting. Our novel method, PLM-interact, goes beyond a single protein, jointly encoding protein pairs to learn their relationships, analogous to the next-sentence prediction task from natural language processing. This approach provides a significant improvement in performance: Trained on human-human PPIs, PLM-interact predicts mouse, fly, worm, E. coli and yeast PPIs, with 16-28% improvements in AUPR compared with state-of-the-art PPI models. Additionally, it can detect changes that disrupt or cause PPIs and be applied to virus-host PPI prediction. Our work demonstrates that large language models can be extended to learn the intricate relationships among biomolecules from their sequences alone.

bioinformatics↗

bioRxiv2021

Using topic modeling to detect cellular crosstalk in scRNA-seq

Cell-cell interactions are vital for numerous biological processes including development, differentiation, and response to inflammation. Currently most methods for studying interactions on scRNA-seq level are based on curated databases of ligands and receptors. While those methods are useful, they are limited to our current biological knowledge. Recent advances in single cell protocols have allowed for physically interacting cells to be captured and as such we have the potential to study interactions in a complimentary way without relying on prior knowledge. We introduce a new method for detecting genes that change as a result of interaction based on Latent Dirichlet Allocation (LDA). We apply our method to synthetic datasets to demonstrate its ability to detect genes that change in an interacting population compared to a reference population. Next, we apply our approach to two datasets of physically interacting cells and identify genes that change as a result of interaction, examples include adhesion and co-stimulatory molecules which confirm physical interaction between cells. For each dataset we produce a ranking of genes that are changing in subpopulations of the interacting cells. In addition to genes, discussed in the original publications we highlight further candidates for interaction in the top 100 and 300 ranked genes. Lastly, we apply our method to a dataset generated by a standard droplet based protocol, not designed to capture interacting cells and discuss its suitability for analysing interactions. We present a method that streamlines the detection of interactions and does not require prior clustering and generation of synthetic reference profiles to detect changes in expression. Author summaryWhile scRNA-seq research is a dynamic area, progress is lacking when it comes to developing methods that allow analysis of interaction that is independent of curated resources of known interacting pairs. Recent advances of sequencing protocols have allowed for interacting cells to be captured. We propose a novel method based on LDA that captures changes in gene expression as a result of interaction. Our method does not require prior information in the form of clustering or generation of synthetic reference profiles. We demonstrate the suitability of our approach by applying it to synthetic and real datasets and manage to capture biologically interesting candidates of interaction.

bioinformatics↗