Week 7: The End ?
This week I worked on the model to assess phenotypes, and started off with the new data that had just got in. however The data that we got ended up not being of high enough depth, and after many, many hours of attempting to split classify and resplit bins using gordian we finally figured this out.
This week I also finished up creating a generalized linear model that would predict phenotypes based on the non-synonymous changes. However this ended up being a missplazed venture, as snpeff would spit out too many false positives for the model, and as I found this out on wednesday I am going to be moving the focus of my poster more towards the script that I created to predict paratpe configurations.
So for the last few days of the week I spent most of my time combining the vcfs from old and the new samples that I had assessed. Then when I started to run the program I realized that there wa a bias towards a high N and B count, so I had to work in a normalization step. This all being said even by the end of the week, even after my poster was created, the script is not perfect, and although it gets the configuration in the top four consistently, the correct paratype configuration is not necessary at the top.
This week was the last working week of the program, but I plan on continuing working on the project during the semester.