Biology subjects

Rutherford, S. T.

Publications and source records attributed to Rutherford, S. T..

4 recordsLinked to original sources

bioRxiv2026

TxConformal: Controlling False Discoveries in AI-Driven Therapeutic Discovery

Artificial Intelligence (AI) is transforming therapeutic discovery by scoring a large set of promising candidates and prioritizing a shortlist for further investigation. Quantifying the reliability of AI scores and preventing false positives among selected candidates is key to the efficiency of the discovery process. Conformal prediction (CP) has emerged as a popular tool for guiding such prioritization, especially via the conformal selection framework to control false discovery rates (FDR) in selecting top-ranked candidates under distributional shift1, 2. However, deploying these advances in real-world therapeutic discovery remains challenging: distribution shifts are difficult to quantify and correct in high-dimensional biomedical data, and practical workflows often require flexible error metrics. Here, we present TO_SCPLOWXC_SCPLOWCO_SCPLOWONFORMALC_SCPLOW, a general framework for trustworthy decision making when building shortlists using AI scores. TO_SCPLOWXC_SCPLOWCO_SCPLOWONFORMALC_SCPLOW adjusts for distribution shift by balancing the hidden representations in AI models and then provides confidence measures for true discoveries of target biological properties. These confidence measures, interpretable as p-values, can be used in conjunction with statistical multiple testing procedures to derive selection decisions with limited false positives or to estimate the errors in given selection decisions. TO_SCPLOWXC_SCPLOWCO_SCPLOWONFORMALC_SCPLOW controls the false positive rate in six real-world tasks spanning various therapeutic discovery stages, modalities, and AI models with realistic data splits. When selecting promising combinatorial genetic perturbations, TO_SCPLOWXC_SCPLOWCO_SCPLOWONFORMALC_SCPLOW nearly halves false-positive selections compared to baseline methods, substantially reducing unnecessary experimental costs by tens of thousands of dollars. When selecting stable protein structures under mutant shifts, TO_SCPLOWXC_SCPLOWCO_SCPLOWONFORMALC_SCPLOW identifies about 10 times more proteins than baseline methods at stringent thresholds when running at a target FDR level of 10%, recovering over 90% of valuable candidates that baseline methods miss due to unaccounted distribution shifts. Furthermore, we demonstrate that TO_SCPLOWXC_SCPLOWCO_SCPLOWONFORMALC_SCPLOW robustly supports various alternative error metrics suitable for resource-constrained settings. Finally, in a prospective fixed-budget virtual screening campaign for novel antibiotic discovery, TO_SCPLOWXC_SCPLOWCO_SCPLOWONFORMALC_SCPLOW predicted false positives in close agreement with experimental outcomes, with substantial improvements over simple baselines.

bioinformatics↗

bioRxiv2024

A high-throughput phenotypic screen combined with an ultra-large-scale deep learning-based virtual screening reveals novel scaffolds of antibacterial compounds

The proliferation of multi-drug-resistant bacteria underscores an urgent need for novel antibiotics. Traditional discovery methods face challenges due to limited chemical diversity, high costs, and difficulties in identifying structurally novel compounds. Here, we explore the integration of small molecule high-throughput screening with a deep learning-based virtual screening approach to uncover new antibacterial compounds. Leveraging a diverse library of nearly 2 million small molecules, we conducted comprehensive phenotypic screening against a sensitized Escherichia coli strain that, at a low hit rate, yielded thousands of hits. We trained a deep learning model, GNEprop, to predict antibacterial activity, ensuring robustness through out-of-distribution generalization techniques. Virtual screening of over 1.4 billion compounds identified potential candidates, of which 82 exhibited antibacterial activity, illustrating a 90X improved hit rate over the high-throughput screening experiment GNEprop was trained on. Importantly, a significant portion of these newly identified compounds exhibited high dissimilarity to known antibiotics, indicating promising avenues for further exploration in antibiotic discovery.

bioinformatics↗

bioRxiv2021

Spatiotemporal organization of BamA governs the pattern of outer membrane protein distribution in growing Escherichia coli cells

The outer membrane (OM) of Gram-negative bacteria is a robust protective barrier that excludes major classes of antibiotics. The assembly, integrity and functioning of the OM is dependent on {beta}-barrel outer membrane proteins (OMPs), the insertion of which is catalyzed by BamA, the core component of the {beta}-barrel assembly machine (BAM) complex. Little is known about BamA in the context of its native OM environment. Here, using high-affinity fluorescently-labelled antibodies in combination with diffraction-limited and super-resolution fluorescence microscopy, we uncover the spatial and temporal organization of BamA in live Escherichia coli K-12 cells. BamA is clustered into ~150 nm diameter islands that contain an average of 10-11 BamA molecules, in addition to other OMPs, and are distributed throughout the OM and which migrate to the poles during growth. In stationary phase cells, BamA is largely confined to the poles. Emergence from stationary phase coincides with new BamA-containing islands appearing on the longitudinal axis of cells, suggesting they are not seeded by pre-existing BamAs but initiate spontaneously. Consistent with this interpretation, BamA-catalyzed OMP biogenesis is biased towards non-polar regions. Cells ensure the capacity for OMP biogenesis is uniformly distributed during exponential growth, even if the growth rate changes, by maintaining an invariant density of BamA-containing OMP islands (~9 islands/m2) that only diminishes as cells enter stationary phase, the latter change governing what OMPs predominate as cells become quiescent. We conclude that OMP distribution in E. coli is driven by the spatiotemporal organisation of BamA which varies with the different phases of growth. SignificanceThe integrity and functioning of the outer membrane (OM) of Gram-negative bacteria depends on the {beta}-barrel assembly machinery (BAM). Although the structure and the mechanism of the complex have been widely explored, little information exists about the organization of the BAM complex and how it dictates protein distribution in the OM. Here, we utilized highly specific monoclonal antibodies to study the spatiotemporal organization of BamA, the key component of this complex. We reveal that BAM organization is dynamic and tightly linked to the cells growth phase. We further discover that the rate of BAM facilitated OMP biogenesis is significantly reduced near the poles. In turn, these features govern the biogenesis patterns and the distribution of OMPs on the cell surface.

microbiology↗

bioRxiv2020

Novel inhibitors of E. coli lipoprotein diacylglyceryl transferase are insensitive to resistance caused by lpp deletion

Lipoprotein diacylglyceryl transferase (Lgt) catalyzes the first step in the biogenesis of Gram-negative bacterial lipoproteins which play crucial roles in bacterial growth and pathogenesis. We demonstrate that Lgt depletion in a clinical uropathogenic Escherichia coli strain leads to permeabilization of the outer membrane and increased sensitivity to serum killing and antibiotics. Importantly, we identify the first ever described Lgt inhibitors that potently inhibit Lgt biochemical activity in vitro and are bactericidal against wild-type Acinetobacter baumannii and E. coli strains. Unlike inhibition of other steps in lipoprotein biosynthesis, deletion of the major outer membrane lipoprotein, lpp, is not sufficient to rescue growth after Lgt depletion or provide resistance to Lgt inhibitors. Our data validate Lgt as a novel druggable antibacterial target and suggest that inhibition of Lgt may not be sensitive to one of the most common resistance mechanisms that invalidate inhibitors of downstream steps of bacterial lipoprotein biosynthesis and transport.

microbiology↗