multiomics Integration analysis

[youngbee12]

If I want to collect the genome, transcriptome, proteome, phosphorylated proteome, and acetylated proteome simultaneously, how should I prepare the parameter file? I only learned from your tutorial that you can use protein/RNA and phosphorylation for causal inference at the same time.

[ozgunbabur]

The latest version of CausalPath can read multiple omics measurements from an integrated single data file. Please include a "Feature" column in the data file. This column uses a one-letter code indicating the data type of the row. Use P for phosphopeptide, G for global protein, R for RNA, A for acetylpeptide, and M for methylpeptide. If you come across any problems, let me know.

[youngbee12]

Thank you! This is very helpful!

[youngbee12]

I have another question. The same params and expr mat can run in Windows normally but not in MacOS(M3 chip).

[youngbee12]

And CasualPath now only support expressive type data（pro，phospho， ace， meth， RNA）, not mutation and copy number data？

[ozgunbabur]

Please try deleting the "hgnc.txt" file under the ".panda" directory, and rerun. It is possible that it was not downloaded properly.

Yes, currently there is no support for mutation and copy number data. Adding copy number support is not hard though. It can be a small project.

Mutation support is harder because we need to know the functional effect of each mutation, which is largely unavailable.

[youngbee12]

Thanks! After deleting the "hgnc.txt" file, it works. Do you have any plans to add CNA data? I found a lot of traces of mutation or CNA in your source code.
Best,
Zhicheng Yang

[ozgunbabur]

Support for CNA and mutation data is in the infrastructure of CausalPath, yes. Those are included just in case we need them one day. But they are not used in practice. If you have ideas on how to include them in a causal model, I would love to discuss them.