Qiyun Zhu
@zhuqiyun.bsky.social
Assistant Professor @ ASU | Studying #microbiome, #evolution, #bioinformatics, #multiomics | Developing scikit-bio (https://scikit.bio) | Open-source enthusiast
Link to the full-text article: rdcu.be/eUcOO
Scikit-bio: a fundamental Python library for biological omic data analysis
Nature Methods - Scikit-bio: a fundamental Python library for biological omic data analysis
rdcu.be
December 12, 2025 at 12:38 AM
Link to the full-text article: rdcu.be/eUcOO
See email. Thanks for your interest!
December 12, 2025 at 12:08 AM
See email. Thanks for your interest!
@xrefugee13.bsky.social I might have an idea to "save" Bacillota by measuring phylogenetic rejection. Let me email you.
July 22, 2025 at 8:38 PM
@xrefugee13.bsky.social I might have an idea to "save" Bacillota by measuring phylogenetic rejection. Let me email you.
Good to know! People may say that more observations offer more chances to break a hypothesis (Bacillota's monophyly). Taxon sampling vs. method robustness is always a dilemma...
July 22, 2025 at 6:48 PM
Good to know! People may say that more observations offer more chances to break a hypothesis (Bacillota's monophyly). Taxon sampling vs. method robustness is always a dilemma...
Making Bacillota monophyletic probably isn't easy. I've been thinking that a taxonomic group should be splitted only when there's strong phylogenetic evidence rejecting its monophyly (instead of supporting). This can probably resolve lots of "_A" suffixes.
July 22, 2025 at 4:24 PM
Making Bacillota monophyletic probably isn't easy. I've been thinking that a taxonomic group should be splitted only when there's strong phylogenetic evidence rejecting its monophyly (instead of supporting). This can probably resolve lots of "_A" suffixes.
(2/2) ...omic data augmentation methods, interoperability with e.g., Anndata and PyTorch, cross-release benchmarks, roadmap of development practices, enriched tutorials for non-microbiome omics.
July 17, 2025 at 8:20 PM
(2/2) ...omic data augmentation methods, interoperability with e.g., Anndata and PyTorch, cross-release benchmarks, roadmap of development practices, enriched tutorials for non-microbiome omics.
Our method selects any number of markers that perform as well as or better than currently adopted markers of this number. More is better. More contains less. Paralogs okay. One can tweak it to bias toward well- or pooly-studied microbial genomes.
March 20, 2025 at 5:30 PM
Our method selects any number of markers that perform as well as or better than currently adopted markers of this number. More is better. More contains less. Paralogs okay. One can tweak it to bias toward well- or pooly-studied microbial genomes.