February 25, 2016
Different groups in class will work on fixing bad characters and identifying new protein functions? (I'm confused by this)
As of late when working with the intestine samples, we are getting a mixture of fed and non-fed when clustering even though DEseq has already told us that clustering goes from fed vs. non-fed. Moving forward, we will play with the filter values and use the correction FPKM. <-- this doesn't sound right... talk to Elise and Kathryn. FPKM is the adjusted expression levels for the length of gene.
Remember, the ultimate goal of this class is a term paper with an introduction, methods, and description of found candidate genes explaining what each candidate gene does and why we selected it. Good potential candidate genes include: transcription factors, transporters, and anything in the signaling cascade.
In class today, we looked for genes belonging to smaller clusters. Other than a lot of trial and error and familiarizing myself with the programming and clustering process, I did not have significant success finding genes that belong to small clusters.
Questions to Consider:
- Is a smaller cluster necessarily advantageous considering that we want to find a gene that turns-on a lot of other genes?