DiPeso, L. | Search bioRxiv

bioRxiv2026

The Synthetic Epitope Atlas: High-Throughput Design and Validation of De Novo Antibody-Antigen Complexes

AO_SCPLOWBSTRACTC_SCPLOWDe novo antibody design models lack sufficient training data to reliably generalize. We demonstrate scalable generation of structural training data for machine learning-driven antibody design by linking in silico designs of antibody-antigen complexes to high-throughput experimental binding validation. Using AlphaSeq, a yeast-based platform for measuring protein binding affinities, we measure the affinity and specificity of thousands of de novo "synthetic epitope proteins" (SEPs) designed to bind to VHHs. The resulting Synthetic Epitope Atlas (SEPIA) pairs over 26 million on- and off-target affinity measurements with computationally designed VHH-SEP "pseudo-structures." We validate strong, specific binding for 1,161 pseudo-structures and >75,000 VHH and SEP mutational variants. We show that these pseudo-structures complement existing structural databases and enable ML models to outperform confidence metrics commonly used to rank de novo antibody designs. Taken together, SEPIA establishes a scalable framework for improving de novo antibody design by augmenting sparse structural data with large-scale experimental binding data.

synthetic biology↗

bioRxiv2023

Machine vision reveals micronucleus rupture as a potential driver of the transcriptomic response to aneuploidy

Recent advances in isolating cells based on visual phenotypes have transformed our ability to identify the mechanisms and consequences of complex traits. Micronucleus (MN) formation is a frequent outcome of genome instability, triggers extensive changes in genome structure and signaling coincident with MN rupture, and is almost exclusively defined by visual analysis. Automated MN detection in microscopy images has proved challenging, limiting discovery of the mechanisms and consequences of MN. In this study we describe two new MN segmentation modules: a rapid model for classifying micronucleated cells and their rupture status (VCS MN), and a robust model for accurate MN segmentation (MNFinder) from a broad range of cell lines. As proof-of-concept, we define the transcriptome of non-transformed human cells with intact or ruptured MN after chromosome missegregation by combining VCS MN with photoactivation-based cell isolation and RNASeq. Surprisingly, we find that neither MN formation nor rupture triggers a strong unique transcriptional response. Instead, transcriptional changes appear correlated with small increases in aneuploidy in these cell classes. Our MN segmentation modules overcome a significant challenge with reproducible MN quantification, and, joined with visual cell sorting, enable the application of powerful functional genomics assays to a wide-range of questions in MN biology.

cell biology↗

bioRxiv2019

IRG1 and iNOS act redundantly with other interferon gamma induced factors to restrict intracellular replication of Legionella pneumophila.

Interferon gamma (IFN{gamma}) restricts the intracellular replication of many pathogens, but how IFN{gamma} confers cell-intrinsic pathogen resistance remains unclear. For example, intracellular replication of the bacterial pathogen Legionella pneumophila in macrophages is potently curtailed by IFN{gamma}, but consistent with prior results, no individual genetic deficiency we tested compromised IFN{gamma}-mediated control. Intriguingly, however, we observed that the glycolysis inhibitor 2-deoxyglucose (2DG) partially rescued L. pneumophila replication in IFN{gamma}-treated macrophages. 2DG inhibits glycolysis and triggers the unfolded protein response, but unexpectedly, it appears these effects are not responsible for perturbing the antimicrobial activity of IFN{gamma}. Instead, we found that 2DG rescues bacterial replication predominantly by inhibiting the induction of two key antimicrobial factors, inducible nitric oxide synthase (iNOS) and immune responsive gene 1 (IRG1). Using immortalized and primary macrophages deficient in iNOS and IRG1, we confirm that loss of both iNOS and IRG1, but not individual deficiency in each gene, partially reduces IFN{gamma}-mediated restriction of L. pneumophila. Further, using a combinatorial CRISPR/Cas9 mutagenesis approach, we find that mutation of iNOS and IRG1 in combination with four other genes (CASP11, IRGM1, IRGM3 and NOX2) results in a total loss of L. pneumophila restriction by IFN{gamma} in primary bone marrow macrophages. There are few, if any, other examples in which the complete set of cell-intrinsic factors required for IFN{gamma}-mediated restriction of an intracellular bacterial pathogen have been genetically identified. Our results highlight the combinatorial strategy used by hosts to block the exploitation of macrophages by pathogens.\n\nImportanceLegionella pneumophila is one example among many species of pathogenic bacteria that replicate within mammalian macrophages during infection. The immune signaling factor interferon gamma (IFN{gamma}) blocks L. pneumophila replication in macrophages and is an essential component of the immune response to L. pneumophila and other intracellular pathogens. However, to date, no study has determined the exact molecular factors induced by IFN{gamma} that are required for its activity. We generated macrophages lacking different combinations of IFN{gamma}-induced genes in an attempt to find a genetic background in which there is a complete loss of IFN{gamma}-mediated restriction of L. pneumophila. We successfully identified six genes that comprise the totality of the IFN{gamma}-dependent restriction of L. pneumophila replication in macrophages. Our results clarify the molecular basis underlying the potent effects of IFN{gamma} and highlight how redundancy downstream of IFN{gamma} is key to prevent exploitation of the macrophage niche by pathogens.

immunology↗