Proteomics data

[Dali Jarboui]

Amazing tool, one stop shop for a full analysis, I have few questions regarding proteomics data :

How omics playground handle proteomics data,?

submitting intensities values, seems like omics playground applies normalization? what if I have LFQ data already normalized ? , 

apparently omics playground applies CPM like normalization on proteomics data ?

Is proteomics data treated with the same workflow as transcriptomics ?

will be helpful if you have any information about the workflow applied to proteomics data , and what is the best way to analyze them (is it better to submit already log transformed intensities?...)

thanks for your help

[BigOmics Analytics]

Hello! 

Your correct on all points. Many people have used the Playground successfully for proteomics data. You can either submit raw abundance or the LFQ values. Omics playground normalizes proteomics with CPM and quantile normalization, treated as transcriptomics. We found CPM normalization necessary to downscale the large intensity values. For some info and references please refer to the FAQ on https://omicsplayground.readthedocs.io/en/latest/faq.html

Missing values are treated as zero, or you need to impute "missing" values yourself. DESeq2 and EdgeR were originally meant for RNA-seq but work nicely with proteomics also for missing/zero values. You can restrict to the t-test and limma based methods if you want a more classical approach.

BigOmics Team

PS. We are looking for great R/Shiny developers with bioinformatics background for our team. If you are interested, or know someone, please contact us.