Hani Goodarzi

A high-resolution protein-RNA binding atlas for structure-aware interaction prediction and post-transcriptional regulatory modeling

The paper introduces ecCLIP for high-resolution RBP binding maps and Pythia for structure-aware RNA-protein interaction modeling, improving RNA half-life prediction and variant effect assessment.

The adaptive architecture of tRNA dependencies across physiological tumor environments

The paper maps tRNA dependencies in cancer cells across different environments, finding that tRNA dependencies are environmentally sensitive while some dependencies are shared across cell lines.

A map of RBP-miRNA regulatory connections for deciphering the mechanisms of transcriptome remodeling in cancer

The paper maps regulatory connections between RBPs and miRNAs in cancer, showing RBP-driven regulation contributes to miRNA activity heterogeneity and identifies DDX55 as a key regulator of miR-29-3p.

PIE: Generalizing perturbation effects across unseen perturbations, contexts and datasets

Predicting differentially expressed genes and expression shifts per condition from database embeddings plus per-gene evidence, PIE reaches 1.2–3.2 times the best baseline's AUPRC on unseen contexts, perturbations and datasets.

A massively parallel synthetic gene atlas for learning compact cis-regulatory grammar across cellular contexts

The paper introduces a synthetic gene atlas to study cis-regulatory elements across cell lines, enabling virtual-cell modeling by expanding gene regulatory networks with defined regulatory codes.

Predicting cellular responses to perturbation across diverse contexts with STATE

State, a set-based transformer trained on over 100 million perturbed cells, predicts perturbation responses in new cellular contexts better than mean, linear and deep learning baselines.

APOE4 disrupts the central dogma by arresting neuronal proteome dynamics

This study shows APOE4 disrupts neuronal proteome dynamics by impairing proteostasis, leading to protein accumulation and cellular stress before Alzheimer's disease onset.

BioReason-Pro: Advancing Protein Function Prediction with Multimodal Biological Reasoning

A reasoning LLM combining ESM3 embeddings with GO-GPT term predictions produces protein function annotations that experts rated tie-or-better than UniProt in 79% of cases (SFT model).

Semantic fragment representations for coordinate-free analysis of genomics data

The paper introduces LEAF-1, a fragment-level foundation model for genomics data analysis, showing it can classify cell types and detect cancer from sparse datasets and cell-free DNA.

Back to basics: Observed statistics are sufficient to predict drug responses

The paper introduces Rhaister, a perturbation-response predictor that uses summary statistics to predict drug responses, outperforming virtual-cell models in speed and accuracy.

Recursive mutational robustness in cancer through intra- and inter-genic compensation

The paper identifies intra-genic mutational robustness through alternative splicing in cancer, showing it can complement inter-genic robustness and is associated with worse disease outcomes.

Systematic identification of chromatin organizers as tuners of intratumoral heterogeneity

The paper identifies RNF8 and MIS18A as chromatin organizers that modulate intratumoral heterogeneity in breast cancer, affecting therapeutic sensitivity and metastasis.

Integrative analysis of mRNA stability regulation uncovers a metastasis-suppressive program in breast cancer

The paper identifies a metastasis-suppressive regulatory axis involving RBMS3 and TXNIP by analyzing mRNA stability in breast cancer using a deep-learning framework.

Multiplexed mosaic tumor models reveal natural phenotypic variations in drug response within and between populations

This study uses GENEVA to analyze drug response variations in cancer cells, revealing mitochondrial activation's role in KRAS inhibition and EMT as a resistance mechanism.

An intrinsically disordered region mediates RNA-binding selectivity and pro-oncogenic activities of LARP6

This study identifies an intrinsically disordered region in LARP6 that restricts conformational flexibility of the La-module, enabling RNA-binding selectivity and promoting cancer cell viability and invasion.

Systemic hypoxia suppresses solid tumor growth

This study shows systemic hypoxia reduces solid tumor growth across multiple cancer types without resistance, through purine metabolism suppression and can be pharmacologically induced.

Scalable probe-based single-cell transcriptional profiling for virtual cell perturbation mapping and synthetic biology phenotyping

ProPer-seq is a cost-effective, barcode-free single-cell perturbation profiling method that enables scalable analysis of synthetic biology perturbations and reveals transcriptional control principles.

Systematic annotation of orphan RNAs reveals blood-accessible molecular barcodes of cancer identity and cancer-emergent oncogenic drivers

The paper systematically identifies and characterizes cancer-emergent oncRNAs across 32 tumor types, showing they act as blood-accessible molecular barcodes for cancer identity and prognosis.

Orphan Non-Coding RNAs Drive Tumorigenesis and Enable Pre-Diagnostic Detection in HPV-Negative Head and Neck Cancer

This study identifies oncRNAs as drivers of HPV-negative head and neck cancer and shows their potential for pre-diagnostic detection.

A multi-modal cell-free RNA language model for liquid biopsy applications

Exai-1 is a transformer-based model that integrates RNA sequence and expression data to improve cfRNA analysis for disease detection.

Identifying a gene signature of metastatic potential by linking pre-metastatic state to ultimate metastatic fate

This study identifies a gene signature of metastatic potential by linking pre-metastatic states to metastatic outcomes in mouse pancreatic cancer and validates its relevance in human tumors.

Tahoe-x1: Scaling Perturbation-Trained Single-CellFoundation Models to 3 Billion Parameters

The paper introduces Tahoe-x1, a large-scale perturbation-trained single-cell foundation model with 3 billion parameters, achieving state-of-the-art performance in cancer-related tasks.

An image-based CRISPR screen reveals splicing-mediated control of HP1α condensates

The paper uses an image-based CRISPR screen to show splicing regulates HP1α condensates through RNA-chromatin interactions, affecting heterochromatin organization and stress response genes.

Single-nucleus transcriptome atlas of orbitofrontal cortex in amyotrophic lateral sclerosis with a deep learning-based decoding of alternative polyadenylation mechanisms

The paper constructs a single-nucleus transcriptome atlas of the orbitofrontal cortex in ALS and identifies cell-type-specific transcriptional changes, introducing a deep learning model to analyze alternative polyadenylation.

Actionable biological programs to enhance EGFR-targeted therapy response unveiled by single-cell lineage tracing in clinically relevant lung cancer models

This study uses single-cell lineage tracing and RNAseq to identify pre-existing resistant tumor subclones in lung cancer models and their transcriptional programs, revealing mechanisms of therapy resistance.

Discovery of a tRNA-regulatory transcription factor that suppresses breast cancer metastasis

The paper identifies ZZEF1 as a tRNA-specific transcription factor that suppresses breast cancer metastasis by regulating tRNA-LysUUU expression and affecting codon-specific translation.

Reconstructing tumor evolution without chromosomal instability reveals in vivo determinants of therapeutic response

This study develops an in vivo platform called SETs to reconstruct tumor evolution without chromosomal instability, identifying therapeutic response determinants in prostate cancer.

scBaseCount: an AI agent-curated, uniformly processed, and autonomously updated single cell data repository

scBaseCount is a single-cell RNA sequencing database that automates metadata extraction and data processing using AI, offering a large, standardized collection of gene expression data across multiple organisms and tissues.

Tahoe-100M: A Giga-Scale Single-Cell Perturbation Atlas for Context-Dependent Gene Function and Cellular Modeling

A 100-million-cell atlas: 50 cancer lines pooled as spheroid villages, treated with 379 drugs at several doses, profiled by combinatorial-barcoding scRNA-seq and embedded with a 10-dimensional scVI model.

Genome modeling and design across all domains of life with Evo 2

A 40B-parameter DNA model trained on 9.3 trillion tokens scores coding and noncoding variants zero-shot, exposes interpretable sequence features, and writes chromosome-scale sequences.