-
Notifications
You must be signed in to change notification settings - Fork 0
Log
Today we initiated Canu, and we are currently waiting for the results from the batch. We also successfully used FASTQC to receive the quality scores for the Illumina reads, the results visual results of which will soon be uploaded to enable reading. The most difficult part of today was getting Trimmomatic to work, where we first had some issues understanding the input and output files needed and also the different necessary parameters. After these issues were resolved, we could not get the code to work. Anna has so graciously agreed to read our code and help us. Next entry we will hopefully understand what went wrong and learn from it.
Today we finally got Trimmomatic to work, it turns out that there was not enough input when it came to the illuminaclip portion of the code, where we initially only had the "fastaWithAdapters" file and needed to add the "seed mismatches", the "palindrome clip threshold" and the "simple clip threshold" to get it to work. With both Quast and Prokka it was smooth sailing, and no real issues were incountered. I did however find it quite difficult to figure out which files are appropriate to use for comparison of the annotation, but with a little help from Anna I learned how to properly navigate the NCBI website to find what I was looking for.
Today I revisited the FASTQC files that were generated during the last session only to have found that the results were blank. I redid Trimmomatic and removed the minlen portion of the code and reran FASTQC to find that I got good results this time. I then initiated running the BWA code, which worked, but I got error messages about space saving and have begun rewriting the code to convert it directly into bai files instead. Hopefully this will solve this problem. While BWA was running, I started working a bit with SPAdes. I haven't entirely figured out how it works, but I hope that with some time this task shouldn't be all that difficult.
Today I got SPAdes to work and got my results from the analysis, which actually looked nice, and the same applies to BWA. I also got HTSeq to work. So all in all, everything worked quite smoothly!
Today I only worked with getting DESeq2 to work, which turned out to be quite difficult. I started by using my two generated files from HTSeq (BH and Serum) and tried really hard to get the code to work with this input. However I always got an error message, no matter how much I changed the input. By the end of the session, I started cutting the files into 3 files each (one for each replicate) and hopefully this will work better.
It turns out that splitting the files into 3 files each did not work, what I however found out was the following: "And given that the underscores won't pass R's make.names filter, you might now have [...] as the column name." Through gaining this information, I changed the names of my sample files so they did not contain underscores, and suddenly the script worked perfectly!