search

bioFAM / MOFA2

Multi-Omics Factor Analysis
https://biofam.github.io/MOFA2/
GNU Lesser General Public License v3.0
283 stars 49 forks source link

Strategies using MOFA for integrating proteomic and CyTOF data #125

Open pdimitron opened 1 year ago

pdimitron commented 1 year ago

Hello,

I had a general question about using MOFA. I recently tried using MOFA to explore some proteomic (359 features) and CyTOF (37 features) in a cohort of 89 patient specimens. The data were normalized separately and I'm getting no factors that have explained variance from both assays. I'm pretty new to MOFA but I've been reading that sometimes when one assay has many more features this might affect the weights. I wanted to know if there were alternative strategies I could try to tease out subtler differences in variation.

Thanks! Peter

rargelaguet commented 1 year ago

Hi Peter, sorry for my late reply. The dimensionality of the two data sets looks fine, so it might be a normalisation issue? Can you share a histogram of the two data sets? (might be easier to do so using the MOFA Slack channel, see https://biofam.github.io/MOFA2/contact.html)