contrastive learning

CellContrast: Reconstructing Spatial Relationships in Single-Cell RNA Sequencing Data via Deep Contrastive Learning

CellContrast reconstructs spatial relationships in single-cell RNA-seq data using contrastive learning, outperforming existing methods and enabling accurate spatial analysis.

Predicting Enzyme Functions Using Contrastive Learning with Hierarchical Enzyme Structure Information

This paper introduces EnzHier, a machine learning algorithm that improves enzyme function prediction by integrating sequence, structural information, and functional hierarchy, achieving better accuracy than existing tools.

SCREEN: a graph-based contrastive learning tool to infer catalytic residues and assess mutation tolerance in enzymes

SCREEN is a graph neural network that predicts catalytic residues and assesses mutation tolerance in enzymes by integrating structural and functional information.

BioRSP: a method for characterizing enrichment patterns in single-cell embeddings

BioRSP quantifies enrichment patterns of cell subsets in single-cell embeddings using polar scanning, providing interpretable summaries and stability diagnostics.

3D-Mol: A Novel Contrastive Learning Framework for Molecular Property Prediction with 3D Information

3D-Mol predicts molecular properties using 3D structure information by contrastive learning on 20M unlabeled data, outperforming baselines on 5 benchmarks.

Cross-modal Graph Contrastive Learning with Cellular Images

The paper proposes a cross-modal graph contrastive learning method that uses cellular images to improve molecular representation learning for downstream tasks.

Contrastive Learning for Robust Cell Annotation and Representation from Single-Cell Transcriptomics

This paper introduces CELLULAR, a contrastive learning model for scRNA-Seq data that improves embedding generalization and cell annotation performance across multiple datasets.

Functionally Adaptive Structural Basis Sets of the Brain: A Dynamic Fusion Approach

This paper introduces dynamic fusion, a method to link brain structure and function over time, showing it captures dynamic functional processes and detects nuanced structural components.

Path-based reasoning in biomedical knowledge graphs with BioPathNet

BioPathNet improves link prediction in biomedical knowledge graphs by using path-based reasoning, outperforming existing methods and identifying novel drug indications and gene interactions.

Contrastive pre-training for sequence based genomics models

This paper introduces cGen, an unsupervised contrastive pre-training method for sequence-based genomic models, showing it improves performance in tasks like chromatin profiling and gene expression prediction.

Learning context-aware, distributed gene representations in spatial transcriptomics with SpaCEX

SpaCEX generates context-aware gene embeddings from spatial transcriptomics data, enabling improved downstream analyses like disease gene identification and spatial pattern detection.

Dimensionality Reduction and Denoising of Spatial Transcriptomics Data Using Dual-Channel Masked Graph Autoencoder

The paper introduces STMask, a graph self-supervised learning method for spatial transcriptomics data, improving spatial clustering and trajectory inference.

Missing Wedge Completion via Unsupervised Learning with Coordinate Networks

This paper introduces an unsupervised learning approach using coordinate networks to address the missing wedge problem in cryoET, improving reconstruction quality and reducing runtime.

CLEAN-Contact: Contrastive Learning-enabled Enzyme Functional Annotation Prediction with Structural Inference

The paper introduces CLEAN-Contact, a contrastive learning framework for enzyme function prediction using amino acid sequences and contact maps, showing improved accuracy over existing methods.

ProteinCLIP: enhancing protein language models with natural language

ProteinCLIP refines pre-trained protein language model embeddings using contrastive learning between amino acid sequences and functional text, achieving state-of-the-art results in protein tasks.

MolLM: A Unified Language Model for Integrating Biomedical Text with 2D and 3D Molecular Representations

MolLM integrates 2D and 3D molecular representations with biomedical text using a unified language model, improving performance on downstream tasks through contrastive learning.

Scalable sequence-informed embedding of single-cell ATAC-seq data with CellSpace

CellSpace provides a scalable sequence-informed embedding for scATAC-seq data, capturing cell subpopulations and transcription factor activity while mitigating batch effects.

Spatial domains identification in spatial transcriptomics by domain knowledge-aware and subspace-enhanced graph contrastive learning

This paper introduces GRAS4T, a graph contrastive learning framework for spatial domain identification in spatial transcriptomics, showing improved performance over existing methods.

Removing direct photocurrent artifacts in optogenetic connectivity mapping data via constrained matrix factorization

The paper introduces PhoRC, a method to remove photocurrent artifacts in optogenetic connectivity mapping data using constrained matrix factorization, improving the accuracy of synaptic connection detection.

Learning Discriminative Representations of Superimposed P waves With Weakly-Supervised Temporal Contrastive Learning

This paper introduces S-TCRL, a weakly-supervised framework for learning discriminative representations of superimposed P waves in ECG signals, achieving improved segmentation performance.

Asymmetric Contrastive Objectives for Efficient Phenotypic Screening

This paper introduces asymmetric contrastive objectives for phenotypic screening, improving performance on benchmark datasets and real-world cell screening tasks.

ISMI-VAE: A Deep Learning Model for Classifying Disease Cells Using Gene Expression and SNV Data

This paper introduces ISMI-VAE, a deep learning model that integrates SNV and gene expression data to classify disease cells, showing improved effectiveness and interpretability.

MOL-AE: Auto-Encoder Based Molecular Representation Learning With 3D Cloze Test Objective

The paper introduces a 3D molecular auto-encoder with a new training objective, achieving better performance than existing methods.

TCR clustering by contrastive learning on antigen specificity.

This paper introduces TouCAN, a contrastive learning method for clustering TCR sequences and predicting antigen specificity, showing comparable performance to existing methods.

Guided Diffusion for molecular generation with interaction prompt

This paper introduces InterDiff, a diffusion model guided by atomic interactions to generate molecules with desired ligand-protein interactions, showing improved or comparable binding modes to known drugs.

ProInterVal: Validation of Protein-Protein Interfaces through Learned Interface Representations

This paper introduces a method to validate protein-protein interfaces using learned representations, achieving 0.91 accuracy on a benchmark dataset.

scDTL: single-cell RNA-seq imputation based on deep transfer learning using bulk cell information

scDTL improves scRNA-seq imputation by integrating bulk RNA-seq data through deep transfer learning and domain adaptation.

Gene-language models are whole genome representation learners

This paper introduces a gene-language model that generates whole-genome vector representations from bacterial species, achieving state-of-the-art accuracy in antibiotic resistance prediction.

Effective structure-aware protein alignment via residue-level contrastive learning

CLAlign improves protein alignment by using contrastive learning to generate structure-aware embeddings, outperforming both unsupervised and sequence-based methods.

PPFlow: Target-aware Peptide Design with Torsional Flow Matching

PPFlow designs target-aware peptides using torsional flow matching, achieving state-of-the-art performance in peptide drug generation and optimization.