Asahi Adachi

ORCID: 0000-0003-3205-4081
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Art, Technology, and Culture
  • Legume Nitrogen Fixing Symbiosis
  • Handwritten Text Recognition Techniques
  • Human Motion and Animation
  • Tuberculosis Research and Epidemiology
  • Cell Image Analysis Techniques
  • Soil Carbon and Nitrogen Dynamics
  • Architecture and Computational Design
  • Plant-Microbe Interactions and Immunity
  • Image Processing and 3D Reconstruction
  • Drug Transport and Resistance Mechanisms
  • Computational Drug Discovery Methods
  • Advanced Technology in Applications

Nara Institute of Science and Technology
2022-2025

Sony Computer Science Laboratories
2022-2023

Takeda (Japan)
2023

The human microbiome is closely associated with the health and disease of host. Machine learning models have recently utilized to predict conditions status. Quantifying predictive uncertainty essential for reliable application these microbiome-based prediction in clinical settings. However, quantification such remains unexplored. In this study, we developed a probabilistic model using Gaussian process (GP) kernel function that incorporates microbial community dissimilarities. We evaluated...

10.1093/bioadv/vbaf045 article EN cc-by Bioinformatics Advances 2025-03-11

Summary Plants accommodate diverse microbial communities (microbiomes), which can change dynamically during plant adaptation to varying environmental conditions. However, the direction of these changes and underlying mechanisms driving them, particularly in crops adapting field conditions, remain poorly understood. We investigate root-associated microbiome rice ( Oryza sativa L.) using 16S rRNA gene amplicon metagenome sequencing, across four consecutive cultivation seasons a high-yield,...

10.1101/2024.09.02.610732 preprint EN bioRxiv (Cold Spring Harbor Laboratory) 2024-09-02

Multidrug resistance (MDR1) and breast cancer protein (BCRP) play important roles in drug absorption distribution. Computational prediction of substrates for both transporters can help reduce time discovery. This study aimed to predict the efflux activity MDR1 BCRP using multiple machine learning approaches with molecular descriptors graph convolutional networks (GCNs). In vitro was determined MDR1- BCRP-expressing cells. Predictive performance assessed an in-house dataset a chronological...

10.1208/s12248-023-00853-y article EN cc-by The AAPS Journal 2023-09-12
Coming Soon ...