Kholina, K. | Search bioRxiv

bioRxiv2025

A foundation model for microbial growth dynamics

Microbial growth dynamics contain rich information about microbial populations, which support applications from antibiotic testing to microbiome engineering. However, the high dimensionality of growth data and the scarcity of large, task-specific datasets have limited generalizable modeling analysis across systems. Here, we develop a foundation model for microbial growth dynamics. It is a large-scale, self-supervised representation model trained on [~]370,000 experimental and simulated growth curves spanning diverse microbial species, environmental conditions, and community contexts. The model learns lower-dimensional latent embeddings that capture essential dynamical features of raw growth data and enable accurate reconstruction of these data. The concise representations enhance predictive performance in diverse downstream applications. Using these embedding, we achieve few-shot learning for antibiotic classification and concentration prediction, accurate forecasting of simulated and experimental communities, and inference of total abundance from relative-abundance data. By extracting transferable representations from heterogeneous datasets, our model provides a general framework for analyzing and predicting microbial community dynamics from limited measurements.

synthetic biology↗

bioRxiv2024

FusOn-pLM: A Fusion Oncoprotein-Specific Language Model via Focused Probabilistic Masking

Fusion oncoproteins, a class of chimeric proteins arising from chromosomal translocations, drive and sustain various cancers, particularly those impacting children. Unfortunately, due to their intrinsically disordered nature, large size, and lack of well-defined, druggable pockets, they have been historically challenging to target therapeutically: neither small molecule-based methods nor structure-based approaches for binder design are strong options for this class of molecules. Recently, protein language models (pLMs) have demonstrated success at representing protein sequences with information-rich embeddings, enabling downstream design applications from sequence alone. However, no current pLM has been trained on fusion oncoprotein sequences and thus may not produce optimal representations for these proteins. In this work, we introduce FusOn-pLM, a novel pLM that fine-tunes the state-of-the-art ESM-2 model on fusion oncoprotein sequences. We specifically introduce a novel masked language modeling (MLM) strategy, employing a binding-site probability predictor to focus masking on key amino acid residues, thereby generating more optimal fusion oncoprotein-aware embeddings. Our model improves performance on both fusion oncoprotein-specific benchmarks and disorder prediction tasks in comparison to baseline ESM-2 representations, as well as manually-constructed biophysical embeddings, motivating downstream usage of FusOn-pLM embeddings for therapeutic design tasks targeting these fusions. We have made our model publicly available to the community at https://huggingface.co/ChatterjeeLab/FusOn-pLM.

bioinformatics↗