contrastive learning

An explainable AI latent space of brain dynamics reveals a cerebello-prefrontal signature of schizophrenia symptoms

This paper develops an interpretable AI framework to map brain dynamics onto a latent space, identifying a cerebello-prefrontal signature associated with schizophrenia symptoms.

TaHL-PTM: Post-Translational Modification Prediction in Proteins via Target-Hooked Discriminative Fine-Tuning of Decoder-only Protein Language Models

TaHL-PTM improves PTM prediction by addressing tokenization issues in decoder-only PLMs, achieving better performance on residue-level tasks.

UMITIC: An unsupervised framework for the joint characterization of cellular phenotypes and spatial neighborhoods in multiplex and hyperplex immunofluorescence imaging data

UMITIC is an unsupervised framework that characterizes cellular phenotypes and spatial neighborhoods in multiplex imaging data, showing improved accuracy over existing methods.

Unsupervised machine-learning identifies latent pyrenoid states linked to mitotic remodeling defects and CO2-dependent growth

The paper uses unsupervised machine learning to identify pyrenoid morphology mutants in Chlamydomonas reinhardtii, linking defects to CO2-dependent growth and mitotic remodeling.

scDisent: regulatory-aware disentangled representation learning for multi-omic single-cell analysis

scDisent separates regulatory and expression variables in multi-omic single-cell data, improving integration and interpretability.

BindScreen: Protein-Centric Contrastive Learning for Sequence-Based Virtual Screening

BindScreen improves virtual screening by using a protein-centric contrastive learning approach with asymmetric loss and batch design, achieving better performance with fewer computational resources.

DeepPathway: Predicting Pathway Expression from Histopathology Images

DeepPathway predicts pathway expression from H&E images using contrastive learning, validated on prostate cancer and brain tumor datasets.

MultiFlow: coupled flow matching for predicting single-cell multiomic perturbation responses in unseen cellular contexts

MultiFlow predicts single-cell multiomic perturbation responses by jointly modeling gene expression and chromatin accessibility across cellular contexts.

Neural and behavioural manifold dynamics align across interacting individuals

This study identifies low-dimensional neural manifolds that align across interacting individuals during coordinated tasks, revealing a novel mechanism for interpersonal coordination through dynamic neural updates.

Time-resolved operator archetypes characterize dynamical sensitivity during cell-state transitions

This paper introduces scJDO, a framework for analyzing dynamical sensitivity during cell-state transitions, and applies it to iPSC reprogramming to distinguish productive from diverted fates.

A novel benchmark dataset for enzyme function prediction reveals the limitations of state-of-the-art models

This paper introduces a benchmark dataset for enzyme function prediction and shows that current models struggle to distinguish non-functional enzyme variants from functional ones.

ARCHER: Amortized cross-specimen pose estimation for cryo-electron microscopy

This paper introduces ARCHER, a method for cross-specimen cryo-EM pose estimation that generalizes across structures using a contrastive classifier trained on protein structures.

Automating Biomedical Knowledge Graph Construction For Context-Aware Scientific Inference

AutoBioKG constructs context-aware biomedical knowledge graphs to improve scientific inference by encoding environmental conditions and entity attributes, outperforming baselines in multiple benchmarks.

Structure-free, site-resolved contrastive learning extends small-molecule discovery beyond the reach of structure-based modeling

Ptarmigan-1 uses contrastive learning to predict small molecule-protein interactions without requiring 3D structure, outperforming traditional methods on challenging targets.

Conformal Uncertainty Quantification for BayesAge Epigenetic Age Predictions

This paper applies conformal prediction to BayesAge to provide calibrated uncertainty intervals for epigenetic age estimation from DNA methylation data.

CALFP-MHC: Interpretable Pan-Allelic Prediction of Peptide-MHC Binding and Presentation Using Chemically Grounded Fingerprints and Contrastive Learning

CALFP-MHC improves peptide-MHC binding prediction by using chemically grounded fingerprints and contrastive learning, achieving high AUC and PPV across diverse alleles and class imbalance.

Detection of Spatially Aberrant Cells in Spatial Transcriptomics Data by Conformal Prediction

The paper introduces SPADE, a framework that detects spatially aberrant cells in spatial transcriptomics data using variational autoencoders and conformal prediction.

Trustworthy in silico labeling via semantic visual interpretability of image-to-image translation

The paper introduces Mask Interpreter, a method to improve interpretability of image-to-image translation models, validating biological relevance and identifying artifacts in in silico labeling.

Learning Discrete Cell and Niche Codes from Spatial Transcriptomics Using Dual Residual Vector Quantization

SQUINT learns discrete cell and niche codes from spatial transcriptomics data using residual vector quantization, outperforming baselines in identification and cross-section integration.

From Generation to Discrimination: Vision Foundation Models for Synthetic SEM Image Detection

This paper develops a deep learning framework to detect synthetic SEM images generated by GANs, using CLIP ViT-L-14 fine-tuned on real and synthetic datasets.

MAE-UNETR++: Masked Autoencoder Pretraining for 3-D Lung Nodule Segmentation

This paper proposes MAE-UNETR++ for 3-D lung nodule segmentation, showing that MAE pretraining outperforms random initialization and Decathlon transfer learning in low-data scenarios.

A Generalized Supervised Contrastive Learning Framework for Integrative Multi-omics Prediction Models

This paper introduces a supervised contrastive learning framework for multi-omics prediction models, showing improved accuracy over single-omics approaches in type 2 diabetes and high-fat diet studies.

Time-frequency embedding with contrastive pre-training allows sub-second seizure detection

This paper introduces a 3D CNN with a trainable CWT layer and BiCL pre-training for sub-second seizure detection, achieving >95% accuracy on EEG data.

ContrastQA: A Label-guided Graph Contrastive Learning-based approach for protein complex structure quality assessment

ContrastQA improves protein complex structure quality assessment using label-guided graph contrastive learning, achieving lower ranking losses than existing methods on CASP16 datasets.

Pro4S: prediction of protein solubility by fusing sequence, structure, and surface

Pro4S predicts protein solubility by integrating sequence, structure, and surface data, improving accuracy and reducing non-expressed proteins in de novo design.

Fast retrieval of structurally similar antibodies from large sequence databases with AbSLang

AbSLang-search efficiently retrieves structurally similar antibodies from large sequence databases using a contrastive learning model for CDR RMSD prediction.

Mapping disease traits onto spatiotemporal domain landscapes through scalable multi-sample integration of spatial transcriptomics

STUltra integrates spatial transcriptomics data to map disease traits onto spatiotemporal tissue landscapes, outperforming existing methods in scalability and speed.

Move BeTween modAlities (MBTA) employs flow matching to predict single cell data modalities

MBTA uses flow matching to address structural mismatch between molecular modalities, improving cross-modal translation and enabling multimodal analysis in single-cell data.

Assessing Computational Models for Pharmacogenomic Variant Interpretation

This study evaluates computational models for predicting pharmacogenomic variant effects, finding that StructureDCA outperforms other methods in distinguishing functional variants.

Unlocking biological insight from single-cell data with an interpretable dual-stream foundation model

This paper introduces scDMC, a dual-stream foundation model for single-cell data that improves representation learning and achieves state-of-the-art results in cell annotation, clustering, and integration.