Teppei Shimamura

This paper designs PDZ-binding peptides using a deep learning method that guides sequence searches in a shared latent space for different binding objectives.

The paper develops a closed-loop framework for designing receptor-selective cell-penetrating peptides using multi-objective optimization and evaluates its effectiveness through experimental validation.

Mievformer uses masked self-supervised learning to model microenvironmental and cellular heterogeneity in spatial transcriptomics data, outperforming existing methods in niche clustering and real dataset analysis.

This study shows that anti-amyloid beta therapy improves stroke recovery in the presence of cerebral amyloid beta accumulation by resolving myeloid immunity impairment.

The paper explores sustainable life-support systems for space by simulating Earth ecosystems, identifying challenges like moisture and temperature, and highlighting microbial roles in ecosystem stability.

DeepKINET estimates single-cell RNA splicing and degradation rates from scRNA-seq data, outperforming existing methods and revealing cellular heterogeneity in post-transcriptional regulation.

ExDyn is a deep generative model that estimates single-cell transcriptome dynamics influenced by extrinsic factors, showing improved accuracy over existing methods.

The paper identifies dysregulated expanded endocannabinoid system metabolites in ALS and shows FAAH inhibition improves motor neuron degeneration in models.

scMM is a deep generative model that extracts interpretable joint representations and enables cross-modal generation for single-cell multiomics data, improving clustering and data integration.

The paper introduces BALSAMICO, a hierarchical Bayesian framework for analyzing microbial communities with environmental factors, and detects bacteria linked to colorectal cancer.

The paper shows that KDM7A and UTX demethylate repressive histone marks to enable NF-κB-dependent inflammatory gene activation in endothelial cells.

CYBERTRACK is a statistical framework for clustering and tracking cell populations in time-series flow cytometry data, showing consistent results with known lymphocyte dynamics.

ENIGMA identifies microbial associations with disease by estimating enterotype-like clusters and environmental effects using a probabilistic model.

The paper introduces a Network of Networks framework and GOSPEL algorithm to model interconnectivity between brain-specific functional networks and identify autism-associated genetic structures.

GIMLET identifies biological modulators in gene regulation by capturing multivariate local dependencies using energy statistics, showing improved performance over existing methods.

This study develops a method to model tumor subclonal progression by linking genetic aberrations to cancer hallmark acquisition using multiregional sequencing data.

This paper introduces BERMUDA, a probabilistic model for analyzing microbial composition and disease, showing improved handling of zero-inflated data and providing efficient learning through an annealing EM algorithm.

The paper introduces phyC, a clustering method for cancer sub-clonal evolutionary trees, and shows it detects clusters related to cancer type or phenotype in real data.

D3M detects differential methylation distributions between case and control groups using Wasserstein metric, showing improved accuracy over existing methods in glioblastoma and lower grade glioma data.