Progress Summary- Final Project (ccRCC)
BIOINFORMATICS PATHWAY FINAL By Croix Mikofsky
Initially I wanted to use a leukemia dataset, but due to time constraints and my personal difficultly using the oligo base package I opted for a less complex experimental design and a dataset that used affy as a base.
Once I began the analysis it went pretty smoothly. My only real difficulties were with transcriptional factor analysis (which I had not done previously) and I decided not to do it. Also my StringDB network was crowded by the number of DEGs I found.
Also I had a little bit of difficultly understanding my findings in the context of renal cancer. Sometimes, I had difficultly explaining from a biological standpoint what it meant for a certain pathway to be enriched.
Challenges were all overcome via correspondence with technical leads and my team.
Summary of work:
Quality control (simpleaffy, affyQCReport, affyPLM)
DEG analysis (limma)
Functional analysis (done for up-reg genes, down-reg, and all)
enrichGO (MF, CC, BP) groupGO, KEGG analysis (enrichKEGG), GSEA analysis, StringDB,
gene concept network (enrichDGN), GEPIA (survival analysis)
Only some of my findings were included in my presentation. I chose to only present results from a few pathways that I could readily explain. One of the more interesting findings was multiple DEGs associated with hypoxia-inducible factors (HIFs). I could connect these genes which regulated the formation of new blood vessels (angiogenesis) directly to the malignant tissue’s tendency to demand more oxygen-rich blood flow from the patient’s body.
In the future I would like to learn how to process data with base packages other than affy, like oligo. I would also really like to work with a really large data set like I had initially planned (might need a stronger CPU, however). Also looking at neurodegenerative diseases would be interesting as well.