Biology subjects

Paschalidis, I. C.

Publications and source records attributed to Paschalidis, I. C..

4 recordsLinked to original sources

bioRxiv2025

CDR-aware masked language models for pairedantibodies enable state-of-the-art bindingprediction

BackgroundTherapeutic antibodies are a leading class of biologics, yet their unique architecture poses challenges for computational modeling. Each antibody comprises paired heavy and light variable domains with conserved framework regions that maintain structure and hypervariable complementarity-determining regions (CDRs) that directly contact antigens. This functional asymmetry, where CDRs determine binding specificity while frameworks provide scaffolding, suggests that region-aware training strategies could yield superior representations. Existing protein language models treat all regions uniformly, potentially missing critical features present in CDRs. MethodsWe developed a region-aware pretraining strategy for paired variable domain sequences using two protein language models: a 3 billion parameter model (ESM2) and a compact 600 million parameter model (ESM C). We compared three masking approaches: uniform whole-chain masking, CDR-focused masking, and a hybrid strategy. Final models were trained on over 1.6 million paired antibody sequences and evaluated on binding affinity datasets with over 90,000 antibody variants across six antigens, including single-mutant panels and combinatorial libraries. ResultsHere we show that CDR-focused training produces embeddings with superior predictive performance for antibody-antigen binding. Our approach achieves up to 27% improvements in binding affinity prediction compared to benchmarked antibody models. Remarkably, training exclusively on paired sequences proves sufficient; pretraining on billions of unpaired sequences provides no measurable benefit. Our compact model matches or exceeds larger antibody-specific baselines. ConclusionsThese findings establish that prioritizing paired sequences with CDR-aware supervision over scale and complex training schemes achieves both computational efficiency and predictive accuracy, providing a practical framework for next generation antibody language models.

bioinformatics↗

bioRxiv2023

ITNR: Inversion Transformer-based Neural Ranking for Cancer Drug Recommendations

Personalized drug response prediction is an approach for tailoring effective therapeutic strategies for patients based on their tumors genomic characterization. The current study introduces a new listwise Learning-to-rank (LTR) model called Inversion Transformer-based Neural Ranking (ITNR). ITNR utilizes genomic features and a transformer architecture to decipher functional relationships and construct models that can predict patient-specific drug responses. Our experiments were conducted on three major drug response data sets, showing that ITNR reliably and consistently outperforms state-of-the-art LTR models. HighlightsO_LIThe proposed framework is a transformer-based model to predict drug responses using RNAseq gene expression profile, drug descriptors and drug fingerprints. C_LIO_LIITNR utilizes a Context-Aware-Transformer architecture as its scoring function that ensures the modeling of inter-item dependencies. C_LIO_LIWe introduced a novel loss function using the concept of Inversion and Approximate Permutation matrices. C_LIO_LIOur computational results indicated that our method leads to substantially improved performance when compared to the baseline methods across all performance metrics, which can lead to selecting highly effective personalized treatment. C_LI

bioinformatics↗

bioRxiv2022

Prediction of representative phenotypes using multi-output subset selection

The interpretation of complex biological datasets requires the identification of representative variables that describe the data without critical information loss. This is particularly important in the analysis of large phenotypic datasets ("phenomics"). We introduce Multi-Attribute Subset Selection (MASS), an algorithm which separates a matrix of phenotypes (e.g., yield across microbial species and environmental conditions) into predictor and response sets of conditions. Using mixed integer linear programming, MASS expresses the response conditions as a linear combination of the predictor conditions, while simultaneously searching for the optimally descriptive set of predictors. We applied the algorithm to three microbial datasets and identified environmental conditions that predict phenotypes under other conditions, providing biologically interpretable axes for strain discrimination. MASS could be used to reduce the number of experiments needed to identify species or to map their metabolic capabilities. The generality of the algorithm allows addressing subset selection problems in areas beyond biology.

systems biology↗

bioRxiv2018

Designing metabolic division of labor in microbial communities

Microbes face a tradeoff between being metabolically independent and relying on neighboring organisms for the supply of some essential metabolites. This balance of conflicting strategies affects microbial community structure and dynamics, with important implications for microbiome research and synthetic ecology. A \"gedanken experiment\" to investigate this tradeoff would involve monitoring the rise of mutual dependence as the number of metabolic reactions allowed in an organism is increasingly constrained. The expectation is that below a certain number of reactions, no individual organism would be able to grow in isolation, and cross-feeding partnerships and division of labor would emerge. We implemented this idealized experiment using in silico genome-scale models. In particular, we used mixed integer linear programming to identify tradeoff solutions in communities of Escherichia coli strains. The strategies we found reveal a large space of nuanced and nonintuitive metabolic division of labor opportunities, including, for example, splitting the TCA cycle into two separate halves. The systematic computation of possible division of labor solutions for 1-, 2-, and 3-strain consortia resulted in a rich and complex landscape. This landscape displays a nonlinear boundary, indicating that the loss of an intracellular reaction is not necessarily compensated by a single imported metabolite. Different regions in this landscape are associated with specific solutions and patterns of exchanged metabolites. Our approach also predicts the existence of regions in this landscape where independent bacteria are viable, but outcompeted by cross-feeding pairs, providing a possible incentive for the rise of division of labor.

systems biology↗