Is CellRank 2 appropriate for public PBMC (steady/near-steady) datasets?

[lilyzhou222]

Hi CellRank team,

I’d like to apply CellRank 2 to public PBMC datasets (e.g., 10x 3′ v2/v3; mostly steady or near-steady peripheral blood). Given that PBMC systems often show weak transcriptional dynamics, I’m seeking guidance on whether CellRank 2 is appropriate in this context and how to configure it responsibly.

[Marius1311]

Hi @lilyzhou222, if the dynamics in your system are not very pronounced, it's probably better to focus on discrete structure in your data, i.e. clustering and matrix factorization/gene module approaches, rather than CellRank analysis.

[lilyzhou222]

Thank you very much for your helpful advice! I’ll look into clustering and gene module approaches as you suggested.

…

------------------ 原始邮件 ------------------ 发件人: "theislab/cellrank" ***@***.***>; 发送时间: 2025年9月22日(星期一) 晚上10:39 ***@***.***>; ***@***.******@***.***>; 主题: Re: [theislab/cellrank] Is CellRank 2 appropriate for public PBMC (steady/near-steady) datasets? (Discussion #1270) Hi @lilyzhou222, if the dynamics in your system are not very pronounced, it's probably better to focus on discrete structure in your data, i.e. clustering and matrix factorization/gene module approaches, rather than CellRank analysis. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>

[Marius1311]

great! Here's one of my favourite tools for gene module anlaysis: https://github.com/YosefLab/Hotspot