I analyzed bulk RNA-seq data using DEseq2, to find differences between primary colorectal cancer and liver metastasis. The results I got for the upregulated genes in the liver metastatic samples make sense, but most of the genes in the primary samples are known in literature as pro-metastatic and tumor migration.

I thought of some possible explanations:

1) Some elements that hold a metastatic boosting potential is already highly expressed in primary samples, but there are some underlying key pathways that might translate this expression into actual migration.

2) pro-metastatic genes are also important for the early development of primary tumors, but not the opposite.

What do you think? what could be the reason for such results ? my code is just fine, and I payed attention to the DEseq log2 results order.