large language model

Prediction of exosomal miRNA-based biomarkers for liquid biopsy

The paper develops models to predict exosomal miRNAs using alignment-based and AI approaches, achieving improved performance with an ensemble method.

Leveraging Large Language Models for Redundancy-Aware Pathway Analysis and Deep Biological Interpretation

MAPA uses large language models to reduce redundancy in pathway analysis and improve biological interpretation of multi-omics data.

Moremi Bio Agent: Leveraging Agentic Large Language Model for the Discovery of Broad-Spectrum Antibiotics for Enterobacteriaceae

This study uses an agentic LLM to design and evaluate 1,002 broad-spectrum antibiotic candidates targeting Enterobacteriaceae, with 774 passing ADMET benchmarks and 391 showing moderate binding to dual targets.

Large Language Models Can Extract Metadata for Annotation of Human Neuroimaging Publications

This paper demonstrates that large language models can effectively extract metadata from neuroimaging publications with performance comparable to human annotators.

Improving Automated Deep Phenotyping Through Large Language Models Using Retrieval Augmented Generation

RAG-HPO improves HPO term assignment using retrieval-augmented generation, achieving higher precision and recall than existing tools in phenotypic analysis of clinical text.

Accelerating Insight Discovery in Large Biomedical Text with Scalable Processing Framework

The paper introduces Quanta, a hybrid system combining large language models with evaluation metrics to improve accuracy and efficiency in processing biomedical text for dermatology.

Evaluating scientific theories as predictive models in language neuroscience

This paper introduces QA encoding models that convert qualitative language theories into interpretable brain response predictors, outperforming baselines in fMRI and ECoG data.

CLM-access: A Specialized Foundation Model for High-dimensional Single-cell ATAC-seq analysis

CLM-access is a foundation model for scATAC-seq data that improves batch correction, cell type annotation, and multi-modal integration compared to existing methods.

BioScientist Agent: Designing LLM-Biomedical Agents with KG-Augmented RL Reasoning Modules for Drug Repurposing and Mechanistic of Action Elucidation

The paper introduces BioScientist Agent, a framework combining a biomedical knowledge graph, variational graph auto-encoder, reinforcement learning, and LLMs to improve drug repurposing and mechanistic reasoning.

Inferring spatial gene expression from tissue images using large-scale histology foundation model with SpaFoundation

This paper introduces SpaFoundation, a histology foundation model that predicts spatial gene expression from tissue images, showing strong performance in prediction and transferability.

CRISPR-GPT for Agentic Automation of Gene Editing Experiments

CRISPR-GPT automates CRISPR-based gene-editing design and analysis by integrating LLMs with domain expertise, demonstrating effectiveness in gene knockout and activation experiments.

Identification and characterisation of bacterial pathogens through Large Language Model-assisted text mining

This paper uses a large language model to mine biomedical literature and identify bacterial pathogens, resulting in a catalog of 1,222 species with human infection evidence.

Multi-Model LLM Architectures for Personalized Summarization and Relevance Ranking in Biomedical Literature

The paper develops a personalized literature review system using LLMs and hybrid ranking to summarize and rank biomedical literature effectively.

Precision Oncology Through Dialogue: AI-HOPE-RTK-RAS Integrates Clinical and Genomic Insights into RTK-RAS Alterations in Colorectal Cancer

AI-HOPE-RTK-RAS enables dialogue-driven analysis of RTK-RAS alterations in CRC, revealing novel associations and ancestry-specific patterns in mutation prevalence and prognosis.

Can large language models reliably extract human disease genes from full-text scientific literature?

This paper benchmarks three LLMs for extracting gene-disease-phenotype relationships from biomedical texts, finding GPT-4o achieves 88.8% accuracy using zero-shot prompting.

Conserved heavy/light contacts and germline preferences revealed by a large-scale analysis of natively paired human antibody sequences and structural data.

PairedAbNGS pools 58 single-cell bioprojects of natural paired human and mouse antibodies, revealing heavy/light V gene pairing biases beyond usage frequencies and conserved CDR3-FWR2 inter-chain contacts.

Reading papers: Extraction of molecular interaction networks with large language models

The paper uses large language models to extract molecular interaction networks from scientific texts, achieving high precision and recall in protein-protein and gene regulatory network relations.

OmicsNavigator: An auditable scientific partner for scalable hypothesis validation in spatial omics

OmicsNavigator translates spatial omics data into testable biological findings using a large language model for hypothesis validation and annotation.

Decoding the JAK-STAT axis in colorectal cancer with AI-HOPE-JAK-STAT: A conversational artificial intelligence approach to clinical-genomic integration

The paper introduces AI-HOPE-JAK-STAT, a conversational AI platform that integrates clinical and genomic data to analyze JAK-STAT pathway alterations in colorectal cancer, revealing survival and mutational associations across patient subgroups.

DeepResearch_Eco: A Recursive Agentic Workflow for Complex Scientific Question Answering in Ecology

DeepResearch_Eco enhances ecological research by recursively synthesizing scientific literature, integrating more sources per word than conventional methods.

SCassist: An AI Based Workflow Assistant for Single-Cell Analysis

SCassist uses large language models to guide and enhance single-cell RNA sequencing analysis by providing recommendations and interpretations throughout the workflow.

From Mutation to Prognosis: AI-HOPE-PI3K Enables Artificial Intelligence-Agent Driven Integration of PI3K Pathway Data in Colorectal Cancer Precision Medicine

AI-HOPE-PI3K enables real-time, population-aware analysis of PI3K pathway alterations in colorectal cancer using natural language queries, revealing survival differences and ancestry-linked biomarkers.

GAME: Genomic API for Model Evaluation

GAME provides a standardized framework for benchmarking genomics models across tasks, enabling consistent evaluation and community contributions.

AI-Driven Science Communication: Leveraging LLMs and Knowledge Graphs for Seamless Knowledge Exchange

EcoToxFred enables non-technical users to access environmental toxicology data through natural language queries using a conversational AI agent.

A Large Language Model-Powered Map of Metabolomics Research

PubMedBERT embeddings of 80,656 metabolomics abstracts, projected with t-SNE and grouped into 20 GPT-4o mini-labelled topics, map the field's domains and trends (1998 to early 2024), browsable online.

PlantConnectome: knowledge graph encompassing >70,000 plant articles

PlantConnectome extracts 5 million functional relationships from 71,000 plant biology abstracts using large language models and provides a database for gene-metabolite interactions and regulatory networks.

SARITA: A Large Language Model for Generating the S1 Subunit of the SARS-CoV-2 Spike Protein

SARITA is a large language model that generates synthetic SARS-CoV-2 Spike S1 sequences, accurately predicting mutations in future variants like Delta and Omicron.

Top-DTI: Integrating Topological Deep Learning and Large Language Models for Drug Target Interaction Prediction

Top-DTI predicts drug-target interactions by combining topological features and large language model embeddings, outperforming existing methods on benchmark datasets.

Extracting knowledge from scientific texts on patient-derived cancer models using Large Language Models: algorithm development and validation

This study develops and validates LLM-based methods to extract PDCM-related entities from scientific texts, showing soft prompting improves small open LLMs' performance.

AutoPM3: Enhancing Variant Interpretation via LLM-driven PM3 Evidence Extraction from Scientific Literature

AutoPM3 automates PM3 evidence extraction from scientific literature using LLMs, improving variant interpretation for rare diseases.