archive
Every paper Pith has read. Search by title, abstract, or pith.
548 papers in q-bio.QM · page 3
-
MRI multifractality weakens with age and dementia
Multifractal Signatures of Ageing and Dementia Development: A Multifractal Space-Filling Curve Analysis
-
Optimal matching metric stabilizes instance segmentation scores
Maximum Matching Accuracy: An Instance Segmentation Evaluation Metric Utilizing Globally Optimal Matching
-
SAEs reveal features that fire only on hazardous proteins
VFUSE: Virulent Feature Understanding with Sparse autoEncoders
-
Contrastive pass fixes false causal links from embedding similarities
Correlation Is Not Enough: Embedding Human Metadata for Individual Causal Discovery
-
Pack count and timing adjustment shifts 35% of medication trajectories
Adjusted trajectory of medication exposure taking into account the periodicity of dispensations and the number of dispensed packs and comparative analysis on EFEMERIS database
-
MACCS encodings let MLP+TL models flag key drug substructures
A systematic investigation of molecular encoding methods for drug property predictions across neural network and Transformer encoder-based model
-
Multi-agent system generates spine MRI reports from multi-sequence scans
A multi-agent system for spine MRI report generation from multi-sequence imaging
-
Index measures how tightly data pin down each model parameter
Parameter uncertainty in dynamical models: a practical identifiability index
-
Two latent infections at boarding best fit Andes virus cruise outbreak
Cruise Ship-Associated Andes Virus Cluster aboard MV Hondius, 2026: A Stochastic Scenario Analysis
-
Open code reproduces Mader model lactate outputs
MetaboliSim: a Python implementation of the Mader model for dynamic and steady-state simulation of muscular energy metabolism
-
Frequency gating with latent queries improves token aggregation
Frequency-Domain Latent Attention Gating for Cross-Domain Token Aggregation
-
RdRp structure classifies RNA viruses where sequences fail
Structure-guided taxonomic placement of divergent RNA viruses with ViraClass
-
Distance scores predict binding model performance without new labels
CaliPPer: quantifying, predicting and improving AI model performance for binding prediction
-
GraphPath achieves 0.92 AUROC on prostate targeted therapy
TRAPS: Therapeutic Response Analysis via Pathway-informed Stratification
-
Deterministic tool lifts AI accuracy on viral data queries to 99.7%
Deterministic access to global viral sequence data enables robust agentic scientific discovery
-
Generative models show weaker binding on shallow pockets
ShallowBench: Benchmarking Generative Drug Design Models on Shallow-Pocket Targets
-
AI loop enriches selective triple-drug combos for breast cancer
Iterative AI-guided optimisation of selective triple-drug combinations for breast cancer
-
Bi-filtration with p-adic distances lifts genomic accuracy on small datasets
$p$-adic Bi-Filtrations for Topological Machine Learning on Genomic Sequences
-
DNAm alone tops median performance in deconvolution benchmark
On the Promises and Limits of Multi-omics Integration for Deconvolution: The HADACA3 Benchmark
-
Federated SPARQL annotates variants without copying databases
Federated SPARQL querying for genomic variant functional annotation
-
Generative model predicts Alzheimer's thinning from gene data at r=0.94
Cross-scale spatially-aware generative modeling of transcriptomic programs underlying neurodegenerative brain organization
-
DSU-Net hits 0.9421 Dice on mammogram lesion outlines
DSU-Net: An Attention-Enhanced Dense Skip U-Net for Breast Lesion Segmentation in Mammographic Images
-
OT potentials make multimarginal flow matching simulation-free
Multimarginal flow matching with optimal transport potentials
-
New tests show TCR models fail to generalize to unseen epitopes
New Benchmarking Shows Limited Generalization Power of TCR Antigenic Epitope Prediction Models
-
Atomic finding matching scores radiology reports at 97.8% concordance
RadSEM: A Finding-by-Finding Metric for Clinical Consistency in Radiology Reports
-
EpiFormer lifts epitope prediction F1 by over 40 percent
EpiFormer: Learning Antigen-Antibody Interactions for Epitope Prediction via Geometric Deep Learning
-
Probabilistic method predicts VOD severity before transplant
Probabilistic learning to perform pre-onset individualised prediction of disease severity: application to Veno Occlusive Disease
-
Models predict next LC feature elution at 98% accuracy
The Language of Elution: Autoregressive Prediction of the Next Feature in Untargeted LC-HRMS Lipidomics
-
Structure-informed alignment is NP-complete for fixed scores
Structure-Informed Multiple Sequence Alignment: A Formal Model and Hardness Results
-
Agentic PLM calls oracles mid-generation to fix designs
AgentPLM: Agentic Protein Language Models with Reasoning-Augmented Decoding for Protein Sequence Design
-
Graph net predicts PROTAC degradability from protein structure
Structure-Aware Prediction of PROTAC-Mediated Protein Degradability via Graph Neural Networks
-
Splice variants split drug resistance into visible and invisible classes
SpliceBind: Isoform-Aware Prediction of Binding Pocket Druggability
-
Direct entropy models predict neural activity without proving mechanisms
Feature leakage and the identifiability of direct-dependency entropy models of neural activity
-
Reformulated models achieve 5-day COVID forecasts within 10% error
Using Machine Learning to Enhance Hyperparameter Optimization in Pandemic Modeling: Case study of COVID-19 Dynamics in Ghana
-
Canonical coherence hypergraphs recover EEG coupling frequencies better
Hypergraphs from multivariate connectivity: caCOH-based EEG/MEG representation
-
caCoh hypergraphs recover couplings with higher contrast than MSC graphs
Hypergraphs from multivariate connectivity: caCOH-based EEG/MEG representation
-
Modulation-reaction networks link reaction flows to their regulators
Modulation-Reaction Networks
-
Cross-box attention in cryo-EM maps lifts protein pretraining
CryoProt: A Protein Pretraining Framework with Cross-Box Interactions on Cryo-EM Density Maps
-
Hierarchical motifs boost PPI prediction accuracy
Enhancing Protein-Protein Interaction Prediction with Hierarchical Motif-based Multimodal Protein Embedding
-
Deep learning predicts fistula risk from CT before pancreatic surgery
Automated Prediction of Postoperative Pancreatic Fistula Using Preoperative Computed Tomography
-
Harmonized perturbation data boosts compound embeddings
Chem-PerturBridge: a harmonized compendium of small molecule perturbation transcriptomic effects
-
Total femur BMD and BMC show largest effect on hip fracture risk
DXA-Derived Skeletal Phenotypes and Hip Fracture Risk: A Backdoor-Adjusted Causal Analysis
-
Future protein variants prove harder to rank than past data suggests
TadA-Bench: A Million-Variant Benchmark for Future-Round Discovery Toward Agentic Protein Engineering
-
Activity cliffs depend on molecular representation geometry
The Geometry of Activity Cliffs: Representation Dependence and Multi-Scale Characterization of Activity Landscapes
-
Higher BMI and African ancestry raise Luminal B odds
Obesity and Sociodemographic Factors in Luminal Breast Cancer
-
Gaussian model corrects subunit motions in dynamic proteins
Gaussian Mixture Model-Based Focused Refinement for Enhanced Flexible Structure Determination in CryoEM and CryoET
-
Model predicts pancreatic cancer up to 3 years early from routine tests
Digitally enriching a screening population for pancreatic cancer using routine blood-based measures and clinical histories
-
Predator-prey equations produce habituation and number learning
Training Ecosystems: A Computational Approach to Uncovering Learning Behavior in Unconventional Contexts
-
Caudal fin tracking measures fish reactions to objects in sea cages
A Novel Computer Vision Approach for Assessing Fish Responses to Intrusive Objects in Aquaculture
-
Particle filter estimates flu states from partial lattice data
Stochastic network epidemic model and particle filter: General framework and application to influenza in Japan