the LRT in DEseq2 make the similar format results to wald test, but log2FoldChange has another mean which I can't understand. so I wonder if the result of DEseq2 by LRT could be used in SPIA or GSEA directly. If not, how to apply it with SPIA or GSEA? could anyone help me? thank you.
I am not familiar with SPIA.
You can use the adjusted p-values from a Wald or LRT. With the LRT, just make sure you read the information in ?results about how multiple coefficients can be tested at once depending on the
reduced designs and the number of levels of the factors involved.