Biology subjects

Altman, R.

Publications and source records attributed to Altman, R..

9 recordsLinked to original sources

bioRxiv2026

AutoScreen: AI Co-Scientist System for Target Discovery in Functional Genomics

Target discovery in functional genomics remains largely manual and time-consuming, lacking systematic tools for efficient and reproducible gene-level hypothesis generation. We introduce AutoScreen, an AI co-scientist system supporting target discovery through both Pre-screen Design, which constructs perturbation libraries de novo from free-text research descriptions, and Post-screen Analysis, which re-ranks experimental screen hits by integrating statistical scores with biological context. AutoScreen leverages the complementary strengths of multiple specialist agents that perform deep research into multi-modal evidence, information restructuring, parallel searches across >26 biomedical databases, evidence synthesis, and target review to provide transparent, explainable gene prioritization with pipeline provenance. Across 320 genome-scale CRISPR screens as expert-curated benchmarks, AutoScreen achieved a ~19% increase in validated-hit recovery among its top 100 predictions, relative to the strongest agent baseline, and required a 1.2-fold smaller library to recover the same number of hits at the top-500 reference point. AutoScreen reached mean average precision more than two orders-of-magnitude above random baseline. Further, we validated AutoScreen in cancer immune-evasion case studies focusing on natural killer (NK) and T-cell therapeutics. AutoScreen recovered NK-resistance genes that were initially lower-ranked in a leukemia screen, moving MUC1, PDPN, and LRRC15 from raw ranks of 118, 81, 1384 to 5, 44, 659, respectively. In follow-up tumor killing assay with primary human NK cells, individual perturbation validated all three hits successfully. Next, in prospective benchmarking across two cytotoxic T-cell-killing screens, AutoScreen recovered 77.1% of ground-truth hits called by the consensus of two gold-standard analysis pipelines (FDR<0.1), achieving 8% improvement over the strongest general-purpose LLM agent baseline. Finally, we re-analyzed >500 public datasets to construct the AutoScreen Resource Hub, a growing knowledge base of pre-computed, annotated reports for CRISPR screens, RNA-seq differential expression, and gene-level UK Biobank genome-wide association studies. This agentic AI approach enables real-time genomics benchmark construction that are continuously updated, expanded, for evaluating frontier AI co-scientists. Overall, AutoScreen enables AI-powered target discovery to be more auditable, scalable, extensible, and reproducible.

bioengineering↗

bioRxiv2026

Unlocking Sensitive Data with SPHERE in the Age of AI

Sensitive human data underpin discoveries across medicine, biology and the social sciences, yet privacy regulation often prevents sharing them with collaborators or artificial intelligence (AI) systems. We introduce SPHERE, a model-free method that makes sensitive datasets directly usable by AI and shareable for open science as a synthetic twin, while the original records never leave the local environment. Across 33 datasets spanning five scientific domains, SPHERE protects individual privacy against adversarial re-identification attacks while preserving the data's statistical structure: means, variances and correlations are reproduced exactly, effect size and P value in linear statistical analysis is numerically identical, nonlinear machine-learning utility is retained, and each twin is generated in seconds on a laptop. Frontier AI agents running on the twin reach the same scientific conclusions as on the original records. Analyses of the twin reproduce genome- and proteome-wide results at UK Biobank scale and recover the findings of landmark studies across three independent cohorts and consortia. The approach also extends to deep-learning embeddings across language, vision and time-series, with minimal utility loss. We make the Stanford Alzheimer's Disease Research Center cohort openly available for the first time, as a SPHERE twin spanning nine modalities that any registered researcher can analyze without an approval process. We release SPHERE with certification of each twin's privacy and fidelity, and an AI agent that autonomously executes research tasks on sensitive data without ever accessing it. Sensitive datasets that are currently closed to research could thus become routine inputs to open science and AI to enable key discoveries.

bioinformatics↗

bioRxiv2025

Heat Shock-Induced PI(4)P Increase Drives HSPA1A Translocation to the Plasma Membrane in Cancer and Stressed Cells through PI4KIII Alpha Activation

HSPA1A, a major heat shock protein, is known to translocate to the plasma membrane (PM) in response to cellular stress and cancer, where it plays protective roles in membrane integrity and stress resistance. Although phosphatidylinositol 4-phosphate [PI(4)P] is essential in this translocation, the signals that trigger and facilitate HSPA1As movement remain undefined.Given that membrane lipid composition dynamically shifts during stress, we hypothesized that heat shock-induced PI(4)P changes are crucial for HSPA1As PM localization. To test this hypothesis, we investigated the mechanisms driving PI(4)P changes and HSPA1A PM localization under heat shock. Lipidomic analysis, enzyme-linked immunosorbent assay (ELISA), and confocal imaging revealed a rapid PI(4)P increase at the PM post-heat shock, with levels peaking at 0 hours and declining by 8 hours. RNA sequencing and protein quantification indicated no transcriptional increase in PI4KIII alpha, the kinase responsible for PI(4)P synthesis, suggesting an alternative regulatory mechanism. Hypothesizing that heat shock enhances PI4KIII alpha activity, we performed ELISA coupled with immunoprecipitation, confirming a significant rise in PI4KIII alpha activity following heat shock. Functional analyses further demonstrated that RNAi-mediated PI4KIII alpha depletion or pharmacological PI(4)P reduction, using GSK-A1, impairs HSPA1As localization to the PM, confirming that HSPA1A translocation is PI(4)P-dependent. Our findings identify PI4KIII alpha activity as a key regulator of PI(4)P accumulation and subsequent HSPA1A recruitment to the PM in stressed and cancer cells. This lipid-mediated response offers new insights into stress adaptation and potentially modifiable pathways for therapeutic interventions to control HSPA1A function in cancer.

biochemistry↗

bioRxiv2024

Heat-Induced Phosphatidylserine Changes Drive HSPA1A's Plasma Membrane Localization

Heat shock protein A1A (HSPA1A) is a molecular chaperone crucial in cell survival. In addition to its cytosolic functions, HSPA1A translocates to heat-shocked and cancer cells plasma membrane (PM). In cancer, PM-localized HSPA1A (mHSPA1A) is associated with increased tumor aggressiveness and therapeutic resistance, suggesting that preventing its membrane localization could have therapeutic value. This translocation depends on HSPA1As interaction with PM phospholipids, including phosphatidylserine (PS). Although PS binding regulates HSPA1As membrane localization, the exact trigger for this movement remains unclear. Given that lipid modifications are a cancer hallmark, we hypothesized that PS is a crucial lipid driving HSPA1A translocation and that heat-induced changes in PS levels trigger HSPA1As PM localization in response to heat stress. We tested this hypothesis using pharmacological inhibition and RNA interference (RNAi) targeting PS synthesis, combined with confocal microscopy, lipidomics, and western blotting. Lipidomic analysis and PS-specific biosensors confirmed a heat shock-induced PS increase, peaking immediately post-stress. Inhibition of PS synthesis with fendiline and RNAi significantly reduced HSPA1As PM localization, while depletion of cholesterol or fatty acids had minimal effects, confirming specificity for PS. Further experiments showed that PS saturation and elongation changes did not significantly impact HSPA1As PM localization, indicating that the total PS increase, rather than specific PS species, is the critical factor. These findings reshape current models of HSPA1A trafficking, demonstrating that PS is a crucial regulator of HSPA1As membrane translocation during the heat shock response. This work offers new insights into lipid-regulated protein trafficking and highlights the importance of PS in controlling cellular responses to stress.

cell biology↗

bioRxiv2022

An Iron-Calcium-Miro Axis Influences Parkinson Risk and Neurodegeneration

Dysregulated iron or Ca2+ homeostasis has been reported in Parkinsons disease (PD) models. Here we discover a connection between these two metals at the mitochondria. Elevation of iron levels causes inward mitochondrial Ca2+ overflow, through an interaction of Fe2+ with Mitochondrial Calcium Uniporter. In PD neurons, iron accumulation-triggered Ca2+ influx across the mitochondrial surface leads to spatially confined Ca2+ elevation at the outer mitochondrial membrane, which is subsequently sensed by Miro1, a Ca2+-binding protein. A Miro1 blood test distinguishes PD patients from controls and responds to drug treatment. Miro1-based drug screens in PD cells discover FDA-approved T-type Ca2+-channel blockers. Human genetic analysis reveals enrichment of rare variants in T-type Ca2+-channel subtypes associated with PD status. Our results identify a molecular mechanism in PD pathophysiology, and drug targets and candidates coupled with a convenient stratification method.

cell biology↗

bioRxiv2021

CRISPRa screening with real world evidence identifies potassium channels as neuronal entry factors and druggable targets for SARS-CoV-2

Although vaccines for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have been successful, there are no good treatments for those who are actively infected. While SARS-CoV-2 primarily infects the respiratory tract, clinical evidence indicates that cells from sensory organs and the brain are also susceptible to infection. While many patients suffer from diverse neurological symptoms, the viruss neuronal entry remains mysterious. To discover host factors involved in SARS-CoV-2 viral entry, we performed CRISPR activation (CRISPRa) screens targeting all 6000+ human membrane proteins in cells with and without overexpression of ACE2 using Spike-pseudotyped lentiviruses. This unbiased gain-of-function screening identified both novel and previously validated host factors. Notably, newly found host factors have high expression in neuronal and immune cells, including potassium channel KCNA6, protease LGMN, and MHC-II component HLA-DPB1. We validated these factors using replication-competent SARS-CoV-2 infection assays. Notably, the overexpression of KCNA6 led to a marked increase in infection even in cells with undetectable levels of ACE2 expression. Analysis of human olfactory epithelium scRNA-seq data revealed that OLIG2+/TUJ1+ cells--previously identified as sites of infection in COVID-19 autopsy studies-- have high KCNA6 expression and minimal levels of ACE2. The presence of KCNA6 may thus explain sensory/neuronal aspects of COVID-19 symptoms. Further, we demonstrate that FDA-approved compound dalfampridine, an inhibitor of KCNA-family potassium channels, suppresses viral entry in a dosage-dependent manner. Finally, we identified common prescription drugs likely to modulate the top identified host factors, and performed a retrospective analysis of insurance claims of ~8 million patients. This large cohort study revealed a statistically significant association between top drug classes, particularly those targeting potassium channels, and COVID-19 severity. Taken together, the potassium channel KCNA6 facilitates neuronal entry of SARS-CoV-2 and is a promising target for drug repurposing and development.

genomics↗

bioRxiv2019

Deep functional synthesis: a machine learning approach to gene functional enrichment

Gene functional enrichment is a mainstay of genomics, but it relies on manually curated databases of gene functions that are incomplete and unaware of the biological context. Here we present an alternative machine learning approach, Deep Functional Synthesis (DeepSyn), which moves beyond gene function databases to dynamically infer the functions of a gene set from its associated network of literature and data, conditioned on the disease and drug context of the current experiment. Using a knowledge graph with 3,048,803 associations between genes, diseases, drugs, and functions, DeepSyn obtained accurate performance (range 0.74 AUC to 0.96 AUC) on a variety of biological applications including drug target identification, gene set functional enrichment, and disease gene prediction. AvailabilityThe DeepSyn codebase is available on GitHub at http://github.com/wangshenguiuc/DeepSyn/ under an open source distribution license.

systems biology↗

bioRxiv2019

GRep: Gene Set Representation via Gaussian Embedding

Molecular interaction networks are our basis for understanding functional interdependencies among genes. Network embedding approaches analyze these complicated networks by representing genes as low-dimensional vectors based on the network topology. These low-dimensional vectors have recently become the building blocks for a larger number of systems biology applications. Despite the success of embedding genes in this way, it remains unclear how to effectively represent gene sets, such as protein complexes and signaling pathways. The direct adaptation of existing gene embedding approaches to gene sets cannot model the diverse functions of genes in a set. Here, we propose GRep, a novel gene set embedding approach, which represents each gene set as a multivariate Gaussian distribution rather than a single point in the low-dimensional space. The diversity of genes in a set, or the uncertainty of their contribution to a particular function, is modeled by the covariance matrix of the multivariate Gaussian distribution. By doing so, GRep produces a highly informative and compact gene set representation. Using our representation, we analyze two major pharmacogenomics studies and observe substantial improvement in drug target identification from expression-derived gene sets. Overall, the GRep framework provides a novel representation of gene sets that can be used as input features to off-the-shelf machine learning classifiers for gene set analysis.

bioinformatics↗

bioRxiv2017

Population-specific imputation of gene expression improves prediction of pharmacogenomic traits for African Americans

Genome-wide association studies (GWAS) are useful for discovering genotype-phenotype associations but are limited because they require large cohorts to identify a signal, which can be population-specific. Mapping genetic variation to genes improves power, and allows the effects of both protein coding variation as well as variation in expression to be combined into \"gene level\" effects.\n\nPrevious work has shown that warfarin dose can be predicted using information from genetic variation that affects protein coding regions. Here, we introduce a method that improves the predicted dose by integrating tissue-specific gene expression. In particular, we use drug pathways and expression quantitative trait loci knowledge to impute gene expression--on the assumption that differential expression of key pathway genes may impact dose requirement. We focus on 116 genes from the pharmacokinetic (PK) and pharmacodynamic (PD) pathways of warfarin within training and validation sets comprising both European and African-descent individuals. We build gene-tissue signatures associated with warfarin dose, and identify a signature of eleven gene-tissue pairs that significantly augment the International Warfarin Pharmacogenetics Consortium dosage-prediction algorithm in both populations. Our results demonstrate that imputed expression can improve dose prediction, in a population-specific manner.

genomics↗