Entering edit mode
5.1 years ago
rahel14350
▴
40
Dear all, have you had any experience to apply machine learning algorithms on RNAseq output (Gene set approached not sample base approaches)? What is the packages or program you used for it? I found this MLSeq package that help to apply machine learning approaches on RNAseq data based on patient sample data. But I want first to do classification and ranking on the differentialy expressed genes from RNAseq data analysis. and to address this question then I want to know is it correct to use the DEGs as the test-set and the background genes as the train-set? Many thanks in advance Kind Regards, Rahel
Its not clear what exactly you want to do ?
There are bunch of machine learning algorithms out there that can be applied on RNA-Seq data. But what are you looking for ? classification of RNA samples based on gene expression profiles ? or something else ? You have three groups, so its a multiclass classification problem.
What I want to do is classification and ranking of differentialy expressed genes (e.g. geneset-based approaches using SVM or RF). in MLSeq is mostly based on patient sample classification (train and test set) and I am not sure if I can use it for the gene set ....Can you please let me know which packages can be used for gene classification using SVM, RF or ANN? Many thanks in advance.
You are not still clear about what you want, anyway you may find these packages helpful: ClassifyR and caret