Biology subjects

Culka, M.

Publications and source records attributed to Culka, M..

4 recordsLinked to original sources

bioRxiv2025

Predicting specificity of TCR-pMHC interactions using machine learning and biophysical models

Understanding the mechanism of T-cell activation and T-cell receptor (TCR) discrimination of MHC-presented epitope peptides (pMHCs) remains an open problem. Machine learning (ML)-based prediction of TCR specificity has gained considerable recent attention. However, the capacity of current models to generalize to peptides unseen during training is currently unknown. Here, we use a proprietary cancer-patient data set that profiles TCR binding to novel regions of peptide space to show that peptide generalization remains an unsolved problem. Specifically, we show that while ML methods have demonstrable utility in predicting TCR specificity for known peptides, they fail to generalize to novel peptides. We also show that physics-based methods utilizing classical energy functions outperform ML methods when predicting TCR binding to novel peptides but underperform them on known peptides. In light of these observations, we develop a new ML model that leverages general knowledge acquired by protein foundation models to achieve better or comparable performance than either ML or biophysical methods on both in- and out-of-distribution TCR-pMHC specificity prediction. We furthermore analyze model performance as a function of distance of TCR sequence specificity between the training and test sets to quantitatively characterize the generalization potential of any given TCR-pMHC model. Our analysis sheds light on the status of modeling TCR-pMHC interactions and suggests new paths forward for continued method development and data acquisition.

immunology↗

bioRxiv2024

Tricked by Edge Cases: Can Current Approaches Lead to Accurate Prediction of T-Cell Specificity with Machine Learning?

The ability to predict T cell receptor (TCR) specificity from sequence could transform immunotherapy, vaccine development, and our understanding of immune recognition. While machine learning approaches have shown promise, progress is limited by the quality of training data and underlying modeling assumptions. Historically, equilibrium binding assays using multimeric pMHCs have been a dominant source of data, but these assays often conflate high-affinity binding with true functional specificity, introducing noise into predictive models. Here, we critically examine two commonly discussed ideas in the field: that TCR specificity prediction can be separated from functional activation modeling, and that unsupervised sequence-based approaches can generalize across diverse antigen contexts. We introduce a cell-based assay for directly quantifying TCR-pMHC binding kinetics using monomeric ligands, while simultaneously assessing early activation via CD3{zeta} phosphorylation. These kinetic parameters provide a mechanistic basis for specificity that avoids the artifacts of equilibrium-based measurements. We propose a predictive modeling framework that integrates biophysical measurements with machine learning, and outline strategies for generating high-throughput training data to support this approach. Our findings highlight the need for functionally informed, mechanistically grounded models to advance generalizable TCR specificity prediction.

immunology↗

bioRxiv2024

Lifting the curse from high dimensional data: Automated projection pursuit clustering for the variety of biological data modalities

Unsupervised clustering is a powerful machine-learning technique widely used to analyze high-dimensional biological data. It plays a crucial role in uncovering patterns, structure, and inherent relationships within complex datasets without relying on predefined labels. In the context of biology, high-dimensional data may include transcriptomics, proteomics, and a variety of single-cell omics data. Most existing clustering algorithms operate directly in the high-dimensional space, and their performance may be negatively affected by the phenomenon known as the curse of dimensionality. Here, we show an alternative clustering approach that alleviates the curse by sequentially projecting high-dimensional data into a low-dimensional representation. We validated the effectiveness of our approach, named APP, across various biological data modalities, including flow and mass cytometry data, scRNA-seq, multiplex imaging data, and T-cell receptor repertoire data. APP efficiently recapitulated experimentally validated cell-type definitions and revealed new biologically meaningful patterns.

bioinformatics↗

bioRxiv2019

Dinucleoside polyphosphates act as 5'-RNA caps in Escherichia coli

Dinucleoside polyphosphates (NpnNs), discovered more than 50 years ago,1 are pleiotropic molecules present in almost all types of cells.2 It has been shown that their intracellular concentration can under stress conditions increase from the {micro}M to mM range 2,3. However, the cellular roles and mechanisms of action of NpnNs are still speculative4,5. They have never been considered as part of the RNA, even though they have similar chemical structures as already known RNA caps, such as the nicotinamide adenine dinucleotide (NAD)6-8 and 7-methylguanylate cap9. Here, we show that both methylated and non-methylated Npn Ns serve as RNA caps in Escherichia coli (E. coli). NpnNs are excellent substrates for T7 and E. coli RNA polymerases (RNAP) and efficiently initiate transcription. Further, we demonstrate that the E. coli decapping enzyme RNA 5 pyrophosphohydrolase (RppH) is able to remove the NpnNs-cap from the RNA. RppH was, however, not able to cleave the methylated forms of the NpnN-caps, suggesting that the methylation adds an additional layer to the RNA stability regulation. Our work introduces an original perspective on the chemical structure of RNA in prokaryotes and the function of RNA caps. This is the first evidence that small molecules like NpnNs can act in cells via their incorporation into RNA and influence the cellular metabolism.

molecular biology↗