Whoops haven’t updated daily posts in awhile… here’s what you missed in the last 3 weeks:
submitted aquaculture 2025 abstract
had first committee meeting
met with individuals from NOAA, WDFW, King County, UW (both admin and profs), WSP and FHL to discuss career paths and opportunities to look out for
GRFP workshop mentoring!
got obscenely sick for a few days
Okay, here’s where we are now:
October 9th
Added some of my DEG findings and methods to a comparison chart with Giles’ outputs to see if our results are consistent. Once we can confirm everything looks good, I can full deep dive with the interpretation of the results.
Last session of the GRFP workshop with the NSF style review. Really cool to see how diverse the applications are!
October 11th
Met with mac to review the results. Seems like there are differences in Giles’ findings with STAR alignment rates and my HiSAT alignment rates. He trimmed before aligning (I did not), so next step is to go back and trim and re align one sample with hisat to see if it improves alignment rate. If it doesn’t, then we’ll use the feature counts matrix that was generated from STAR alignmened samples to run downstream analysis from there. If it does, then I’ll trim all the samples, re-align them with hisat, and then do the downstream analyis.
This will have to wait until next week though because the Col. of Env. travel fund application is due today, and not the 18th as I originally thought.
October 16th
Re writing the analysis as mentioned on Friday. Haven’t run it yet as I’ll need to go onto campus to get access to raven and my raw data files again without just re downloading it all.
Starting to think about chapter 1 methods. Found the paper Luke talked about in my committee meeting: “The genetic basis of novel trait gain in walking fish”
October 17th
Read the sea robin paper about fish walking genes – super cool!!! The walking sea robins had 3 pectoral fin rays that are highly modified for walking and seperated from the rest of the fin once the sea robin is fully developed. They were able to identify the genes involved through RNA-seq before, during, and after development of these modifications (before/during after the ‘seperation’ of the modified rays). They then used crispr to turn off the gene they identified and then raised hybrid sea robins that had between 2 and 4 modified rays once fully developed! Super cool insights into not only the underlying genes, but how the phenotype is inherited!
TGIT
Was on campus and was able to execute the trim and hisat code. Trimming improved alignment for my selected sample from ~63% to ~67% mapping rate, but still substantially lower than Giles’ mapping rate (~ 75%). Next step to re-run
October 18th
Started re-running deseq2 using Giles’ STAR count data. Got done with the heatmaps and realized I had a sample mislabled which made the treatment vs. control for everything funky. need to edit and re run. Womp womp. Have work so, problem for another day.
October 20th
Finished re running deseq2. Used original hisat parameters of excluding any genes that dont have: at least 8 individuals in treatment OR 8 individuals in control with at least 23 reads (hisat median non-zero read count for all genes).
Re ran again with updated parameters of 27 reads (updated/star median non-zero read count for all genes).
Updated comparison stats chart AND comparison DEG chart. results from comparison DEG chart:
