Dec 11st 17th

• Ok now the differential expression profile has been finished by other members in the group. The discussion goes into the next stage.
• About the DE results:

1. DESeq2 gives out much more genes in the last stage (NASH->HCC)! It is much harder to analyze them so filtering-down is needed--perhaps we can consider only the genes enriched in tissue, according to HPA?
2. AVOVA gives quite a lot genes that are differentially expressed. However these results are throughout the whole expression progress from healthy to HCC and cannot be annotated to a certain stage. So they are not as useful as the DESeq2 results, since the research interest here in our study is to differentiate among specific stages.

• Differentially expressed genes have been used to enrich pathways out of GO, KEGG and Onco, using Piano. About the later work:

1. Build pathway network.
2. We may not have time for co-expression network; but that's still an option.
3. The results of Piano can be visualized by Cytoscape (network building).

• Need to validate the results by literature reading. (None of the members in our group has a solid biological background actually. We think that we don't need to focus too much on the biological meaning of the genes/pathways we found; the main aim is just to show the results.)