Ko Abe

mmVelo: A deep generative model for estimating cell state-dependent dynamics across multiple modalities

mmVelo estimates cell state-dependent dynamics across multiple modalities using deep generative modeling, validated on mouse and human data.

Skip-Zeros Variational Inference in the Million-Cell Era of Single-Cell Transcriptomics

UNISON is a scalable framework for single-cell transcriptomics that uses skip-zeros variational inference to efficiently analyze large, sparse datasets while preserving interpretability.

scTREND: An annotation-free single-cell time-resolved and condition-dependent hazard model

scTREND is a hazard model that estimates cell-level risk across time and conditions without cell-type annotations, showing improved prediction in simulations and applications to cancer and COVID-19.

scSurv: a deep generative model for single-cell survival analysis

scSurv estimates individual cellular contributions to cancer patient survival using a deep generative model combined with Cox proportional hazards analysis.

A deep generative model for estimating single-cell RNA splicing and degradation rates

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

LineageVAE: Reconstructing Historical Cell States and Transcriptomes toward Unobserved Progenitors

LineageVAE reconstructs historical cell states and transcriptomes from scRNA-seq data to infer unobserved progenitor cell states.

scMM: Mixture-of-experts multimodal deep generative model for single-cell multiomics data analysis.

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.

Hierarchical non-negative matrix factorization using clinical information for microbial communities.

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

Model-based cell clustering and population tracking for time-series flow cytometry data

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: An Enterotype-Like Unigram Mixture Model for Microbial Association Analysis

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

A Latent Allocation Model for the Analysis of Microbial Composition and Disease

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.