2015 Oysterseed Plan and Crab Meeting Notes

Crab Meeting today: shared current status of what we’ve got (FISH546 grace-Cbairdi-transcriptome). Additionally, we had a proteomics meeting today and I got some input on next steps for the DIA: set some stringent thresholds, MS Stats, make some tables, etc. I will make a separate post on what I did (protocol) tomorrow as I work with the data. Today’s post will just include a general outline of what I have and next steps.

Crab Meeting

(Will edit and publish within the next couple of days)

Shared my current status of transcriptome annotation and basic analyses (FISH546 grace-Cbairdi-transcriptome).

Steven and Sam were both surprised by the taxonomy pie chart, specifically that there were ~2500 taxonomic groups identified. (Rscript; (data_set_from_Steven). I will go back and investigate which transcriptome was used, and also make clear that there are two different fasta files: one from the practice assembly using the tiny .fastq files; one from the true assembly on Mox.

I also am going to get some more BLAST jobs running as we wait for the Bioanalyzer to be fixed (after which I can hopefully continue with RNA extractions if results look good).

2015 Oysterseed Proteomics Chat

What I have (finally!): I have some *.elib.peptides.text files, which are the quant files from the DIA.

(Tomorrow’s post will be dedicated to the DIA protocol I used.)

What I will do: I will set a stringent threshold and use MS Stats, take a rough look at the data. Make a table. Emma said she’ll be available to help if (when) I get stuck.

I also need to make sure I can identify the proteome I used. I know where it is, but I’ll include it in tomorrow’s post with the protocol.

Written on December 6, 2018