contrastive learning

This paper introduces f-LENNS, an AI framework that enhances fMRI resolution without retraining, improving spatial fidelity and generalizing to atypical brain function.

This paper introduces Perturbscape, a framework that uses perturbation-induced gene programs to identify disease-relevant genes and pathways, showing higher heritability informativeness than steady-state approaches.

MaxCAT-Net improves skin lesion classification and segmentation with hybrid architecture and calibration, showing performance on benchmark datasets but with preliminary results.

MUTUAL enables cross-modality analysis between transcriptome and chromatin by learning a shared RNA-ATAC space from multiome data without refitting.

The paper evaluates sparse autoencoder features on various embeddings, finding that reproducibility does not guarantee meaningfulness, and proposes reporting features with concept coverage and purity.

The paper examines how auditory tracking changes during wakefulness and propofol anesthesia by analyzing response magnitude, temporal alignment, and anatomical distribution using sEEG data.

LoCoFlow improves trajectory inference in single-cell data by incorporating local geometry and fate uncertainty, achieving better performance than existing methods.

CloneTrast uses contrastive learning to infer T-cell clonal niches from single-cell RNA-seq, identifying NK-like niches linked to poor immunotherapy response across cancers.

This study integrates multi-omic data from mouse and human choroid plexus to identify conserved molecular programs of epithelial function across species.

This paper introduces HARMONIC, a framework that improves cell-cell communication inference in spatial transcriptomics by incorporating histology data, showing it better captures anatomical context.

CohortFM reconstructs continuous molecular trajectories from sparse longitudinal data, revealing biologically meaningful progression patterns across diseases.

This study evaluates cross-cohort generalization of mammography diagnostic models using different image representations and domain-adversarial learning, finding that MedSigLIP embeddings provide the best transferability.

This paper improves histology-based molecular receptor prediction by integrating RNA-seq data during training, showing that gene selection strategy is key to model performance.

Translating cells between mouse, macaque and human scRNA-seq with a CycleGAN, without ortholog mapping, preserved developmental time structure, imputed missing states and produced more realistic synthetic cells than scGAN.

This paper introduces class-conditional conformal prediction to address uneven coverage in multi-class psychiatric neuroimaging, showing improved coverage consistency across diagnostic classes.

CSGDA predicts drug responses across biological heterogeneities by addressing distribution shifts using a cell state-guided graph domain adaptation framework.

Haruka identifies condition-specific and shared spatial domains across tissues and conditions using contrastive learning, demonstrating improved detection of spatially heterogeneous perturbation responses.

SHOT-CCR improves classification of genetic perturbations in Cell Painting datasets by reducing batch effects through biologically guided adversarial training.

NucXplore converts H&E-stained nuclei into 129 explicit features, outperforming classical and deep learning methods in histopathology cohorts and revealing aging and exposure-related cellular changes.

This paper introduces a counterfactual explanation algorithm for graph neural networks to explain patient outcome predictions in patient similarity networks.

OpenCRS is a cardiorespiratory model calibrated with global sensitivity analysis to identify influential parameters and mechanisms during rest and exercise.

scACORN improves single-cell transcriptomic interpretation by combining specialized small language models with context-engineered agent orchestration, achieving higher accuracy and reducing unsupported gene citations.

ImmuneLens links transcriptional states and TCR clonotypes using disentangled multimodal learning to improve antigen-specificity prediction and reveal T cell heterogeneity in immunotherapy.

The paper introduces a diploid genomic foundation model to represent diploid genotypes in genomic language models, improving variant discrimination in compound-heterozygous cases.

The paper predicts cancer treatment response using a MIL-attention framework that analyzes macrophage states from single-cell RNA data, showing improved performance over existing methods.

This paper develops a multimodal framework to align protein sequences with cryo-EM density maps using joint representation learning, achieving strong retrieval performance across species and resolutions.

The paper reanalyzes a proteomic atlas using audited vibe coding to assess if tumors show fetal-like proteomic convergence, finding partial alignment with reverse fetal-to-adult maturation.

This paper improves sickle cell disease morphology analysis by combining a YOLO detector with a DenseNet classifier for accurate detection and classification of red blood cell morphologies in full-scope microscopy images.

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