Quantitative Methods (Biology)
Mathematical and computational methods applied to biological systems.
q-bio.QM · 137 papersBioResearcher: Scenario-Guided Multi-Agent for Translational Medicine
BioResearcher is a multi-agent AI system that uses scenario-guided playbooks and specialized tools to automate complex translational medicine research.
Planar morphometry via functional shape data analysis and quasi-conformal mappings
FDA-QC is a new method for planar shape analysis combining functional data and quasi-conformal maps to capture both boundary and interior features.
Resolving the bias-precision paradox with stochastic causal representation learning for personalized medicine
sMMD, a stochastic causal representation learning method, resolves the bias-precision paradox in personalized medicine, improving individualized treatment effects.
Direct From Darwin: Deriving Advanced Optimizers From Evolutionary First Principles
This paper unifies Fisher's and Wright's evolutionary theories to derive advanced gradient optimizers, showing many existing algorithms are evolutionarily compliant.
When Does Gene Regulatory Network Inference Break? A Controlled Diagnostic Study of Causal and Correlational Methods on Single-Cell Data
This paper diagnoses why causal gene regulatory network inference methods often fail, revealing they excel in clean data but are vulnerable to specific pathologies.
Curated AI beats frontier LLMs at pharma asset discovery
Gosset, an AI platform with curated data, significantly outperforms frontier LLMs in discovering pharma assets, especially niche and preclinical drugs.
FairEnc: A Fair Vision-Language Model with Fair Vision and Text Encoders for Glaucoma Detection
FairEnc is a VLM pretraining method that debiases both vision and text encoders for fair glaucoma detection across diverse patient populations.
Enhancing Cryo-EM Density Map Segmentation in Phenix for Improved Atomic Model Building
PhenixCraft automates cryo-EM atomic model building by integrating AlphaFold, improving segmentation and model accuracy over traditional Phenix methods.
A Zero-Inflated Beta Mixture Model for Marginal Mediation Analysis with Compositional Microbiome Mediators
Introduces ZIBM, a new statistical model for robust mediation analysis of compositional microbiome data, addressing sparsity and heterogeneity.
Modularity Emerges from Action-Functional Constraints in Marine Metabolic Networks: A Biology-Scale Validation of the Network-Weighted Action Principle
This paper validates the Network-Weighted Action Principle, showing marine metabolic networks exhibit significant modularity excess consistent with cost-minimization.
Benchmarking open-source tools for in silico antiviral drug discovery
This paper benchmarks 15 open-source computational tools for antiviral drug discovery, identifying top performers and highlighting data curation.
Tree-Conditioned Edit Flows for Ancestral Sequence Reconstruction
A new tree-conditioned edit-flow model reconstructs ancestral protein sequences, effectively handling insertions and deletions, unlike classical ASR methods.
Graph Neural Network based Hierarchy-Aware Embeddings of Knowledge Graphs: Applications to Yeast Phenotype Prediction
This paper introduces a GNN-based method for hierarchy-aware knowledge graph embeddings, improving yeast phenotype prediction and enabling biological discovery.
ProtDBench: A Unified Benchmark of Protein Binder Design and Evaluation
ProtDBench is a unified, throughput-aware benchmark framework for fair evaluation and comparison of protein binder design methods.
Cusped singularities organize mixed-mode oscillations in mutually inhibitory slow-fast systems
Cusped singularities universally organize mixed-mode oscillations (MMOs) in mutually inhibitory slow-fast neural systems.
Expanding functional protein sequence space using high entropy generative models
High-entropy Boltzmann Machines generate functional proteins and explore vastly larger sequence spaces, better representing evolutionary landscapes.
A-CODE: Fully Atomic Protein Co-Design with Unified Multimodal Diffusion
A-CODE is a fully atomic, unified multimodal diffusion model for protein co-design, achieving superior designability and outperforming existing methods.
Donor-Aware scRNA-seq Benchmarks for IBD Classification
This paper benchmarks donor-aware scRNA-seq methods for IBD classification, showing compartment-aware features improve performance and interpretability.
Dynamic Vine Copulas: Detecting and Quantifying Time-Varying Higher-Order Interactions
Dynamic Vine Copulas (DVC) model time-varying non-Gaussian dependence, detecting and quantifying higher-order interactions missed by traditional methods.
Approaching human parity in the quality of automated organoid image segmentation
This paper introduces a composite AI method combining SAM with a domain-specific tool to achieve near human-level accuracy in automated organoid image segmentation.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.