Grace’s Notebook: Crab Project – What I did for PCSGA 2019

In this post I will detail what I did for PCSGA 2019. I did some new analyses and found some new results from our first assembled transcriptome.


Steven made a BLAST database from the Dinophyceae proteins (Dinophyceae is class under which dinoflagellates are), and did a BLASTx against the first assembled transcriptome (Day 26, infected and uninfected, cold and ambient).

I used this script to separate the transcriptome into putative crab and putative Hematodiniium genes: sep_crab-hemat-genes.Rmd

That script resulted in the following files:

Annotating BLAST outputs

In jupyter notebooks, I annotated the two blastx outputs (crab and Hematodinium) with GOslim terms.

I then used the following R script to create files for creating pie charts for biological process GO slim terms:

The script resulted in the following files:

Creating pie charts

I used the count.csv files to create pie charts in google sheets:

For the talk, I went more in-depth into the “stress response” slice in the crab GOslim pie. I made a table of some crab genes with notes on their names and functions, and links to the uniprot database on those genes:

The ones highlighted are the ones I chose to talk about in the talk.

The talk and slides

Link to final google slides: Crandall_Wed_1645

Link to slidedeck on figshare: Effects_of_Bitter_Crab_Disease_on_the_gene_expression_of_Alaskan_Tanner_Crabs

Thoughts on talk

It was supposed to be 12 minutes, with 3 minutes for questions. I have no idea how much time I took, but I did get 2 clarifying questions at the end of the talk.

It was my first time presenting at a conference, and my slot was on day 2 (Wednesday) at 4:45pm. It was a struggle to stay energized and it was also a struggle to stay calm – I was really nervous!

I think I had good background information and that I explained things fairly well, but I wish I had gone a little more in depth into the genes that we found. I also think that for next time, I need to find ways to combat the nerves because they got a bit in the way of my ability to speak at the beginning – I collected myself well-enough after the third slide or so… but I want to improve!

from Grace’s Lab Notebook

Shelly’s Notebook: Tues. Sept. 24, Geoduck Broodstock Histology

This post is in reference to the histology done on the Fall-Winter 2018-2019 Broodstock conditioned in constant low pH. This is data analysis for the manuscript on pH effect on reproductive development

Past histology analysis

Scoring females with imageJ

Analysis of scores

from shellytrigg

Shelly’s Notebook: Mon. Sept. 23, Geoduck Broodstock Histology and Juv. low pH DMRs

Broodstock histology

Met with Kaitlyn in the am and came up with the following plan for scoring gonad histology:

  1. Females:
    • quantify follicle area
    • quantify egg area
    • calculate egg/follicle ratio
    • calculate follicle/tissue ratio
  2. Males:
    • quantify acini/tissue ratio
    • quantify spermatagonia/spermatid ratio (dark purple to light purple)

Comparing allc DMRs and 5x cov files

  • Steven’s 5x cov files:
    • still unsure about what happened in the strandedness code
    • include bases where MAPQ < 30
  • Allc files generated by methylpy (here):
    • filtered for bases with MAPQ >= 30 (–min-mapq 30 default)
  • Methylkit processBismarkAln default includes a ‘minqual’ filter for MAPQ >= 20
  • Not sure if [DMG pipeline] includes a MAPQ cutoff


  • whatever files we use to validate our DMRs, DMLs, or DMGs, they must be filtered the same way.
    • Otherwise differences may not be apparent when all reads are included…

Some lit on whether to use a MapQ score threshold or not:

from shellytrigg