Protein-protein interactions (PPIs) are important for many biological processes, but predicting them from sequence data remains challenging. Existing deep learning models often cannot generalize to proteins not present in the training set and do not provide uncertainty estimates for their predictions. To address these limitations, we present TUnA, a Transformer-based uncertainty-aware model for PPI prediction. TUnA uses ESM-2 embeddings with Transformer encoders and incorporates a Spectral-normalized Neural Gaussian Process. TUnA achieves state-of-the-art performance and, importantly, evaluates uncertainty for unseen sequences. We demonstrate that TUnA's uncertainty estimates can effectively identify the most reliable predictions, significantly reducing false positives. This capability is crucial in bridging the gap between computational predictions and experimental validation.

MOTIVATION: Antibody therapeutic candidates must exhibit not only tight binding to their target but also good developability properties, especially low risk of immunogenicity. RESULTS: In this work, we fit a simple generative model, SAM, to sixty million human heavy and seventy million human light chains. We show that the probability of a sequence calculated by the model distinguishes human sequences from other species with the same or better accuracy on a variety of benchmark datasets containing >400 million sequences than any other model in the literature, outperforming large language models (LLMs) by large margins. SAM can humanize sequences, generate new sequences, and score sequences for humanness. It is both fast and fully interpretable. Our results highlight the importance of using simple models as baselines for protein engineering tasks. We additionally introduce a new tool for numbering antibody sequences which is orders of magnitude faster than existing tools in the literature. AVAILABILITY AND IMPLEMENTATION: All tools developed in this study are available at https://github.com/Wang-lab-UCSD/AntPack.

How the locus-specificity of epigenetic modifications is regulated remains an unanswered question. A contributing mechanism is that epigenetic enzymes are recruited to specific loci by DNA binding factors recognizing particular sequence motifs (referred to as epi-motifs). Using these motifs to predict biological outputs depending on local epigenetic state such as somatic mutation rates would confirm their functionality. Here, we used DNA motifs including known TF motifs and epi-motifs as a surrogate of epigenetic signals to predict somatic mutation rates in 13 cancers at an average 23kbp resolution. We implemented an interpretable neural network model, called contextual regression, to successfully learn the universal relationship between mutations and DNA motifs, and uncovered motifs that are most impactful on the regional mutation rates such as TP53 and epi-motifs associated with H3K9me3. Furthermore, we identified genomic regions with significantly higher mutation rates than the expected values in each individual tumor and demonstrated that such cancer-related regions can accurately predict cancer types. Interestingly, we found that the same mutation signatures often have different contributions to cancer-related and cancer-independent regions, and we also identified the motifs with the most contribution to each mutation signature.

Transcription factors (TFs) are crucial for regulating cell differentiation during the development of the immune system. However, the key TFs for orchestrating the specification of distinct immune cells are not fully understood. Here, we integrated the transcriptomic and epigenomic measurements in 73 mouse and 61 human primary cell types, respectively, that span the immune cell differentiation pathways. We constructed the cell-type-specific transcriptional regulatory network and assessed the global importance of TFs based on the Taiji framework, which is a method we have previously developed that can infer the global impact of TFs using integrated transcriptomic and epigenetic data. Integrative analysis across cell types revealed putative driver TFs in cell lineage-specific differentiation in both mouse and human systems. We have also identified TF combinations that play important roles in specific developmental stages. Furthermore, we validated the functions of predicted novel TFs in murine CD8+ T cell differentiation and showed the importance of Elf1 and Prdm9 in the effector versus memory T cell fate specification and Kdm2b and Tet3 in promoting differentiation of CD8+ tissue resident memory (Trm) cells, validating the approach. Thus, we have developed a bioinformatic approach that provides a global picture of the regulatory mechanisms that govern cellular differentiation in the immune system and aids the discovery of novel mechanisms in cell fate decisions.