Yaamini’s Notebook: Gonad Methylation Analysis Part 18

Preparing for discoveries

Before I can understand where differentially methylated loci (DML) are located within the C. virginica genome, I first need to identify DMLs! I used this R script to identify DMLs that were at least 50% different between control and treatment (high pCO2) samples (.csv here). Looking at the PCA was interesting, as the clustering was not as tight as I expected.

PCA

Figure 1. Principal Components Analysis of methylated regions in samples.

O2-5 (ambient conditions) are more closely clustered than any of the oysters from treatment conditions. It’s possible that there are organismal differences in methylation responses, or that we just didn’t have a large enough sample size to deal with this variation.

In the last part of the script, I saved my DML information as a BED file. I mimicked Steven’s code to do this. BED files have chromosome ID, start, and stop positions that I can use to pare down information about my DMLs. I can then compare the DML location with other important genomic features using bedtools. The intersect tool seems especially useful. I know we covered bedtools in the 2016 Bioinformatics class, so I’ll review those notes!

// Please enable JavaScript to view the comments powered by Disqus.

from the responsible grad student https://ift.tt/2Jjw3to
via IFTTT

Yaamini’s Notebook: Gonad Methylation Analysis Part 17

Note to self: Always double check things

  • I forgot to change the code in my subset and full sample notebooks so that bismark_methylation_extractor ran on the files I produced instead of those in the dignore folder. I switched the code and everything still works!
  • I thought I double checked what bismark_methylation_extractor outputs needed to be in the .gitignore but I left out several *deduplicated.txt files that were well over 100 MB. My mistake took me three days and one Github issue to figure out. Whoops. Now I know how to use the Github command line, add things to my .gitignore, and effectively undo commits to make Github Desktop happy.

I finished up the methylation extraction, HTML report, and summary report steps! I then started methylKit on the full samples to ensure reproducibility. When I’m finished, I’ll create BEDfiles and start to understand where differentially methylated loci are located and waht the gene functions are.

// Please enable JavaScript to view the comments powered by Disqus.

from the responsible grad student https://ift.tt/2L8F6e1
via IFTTT

Sam’s Notebook:Transposable Element Mapping – Crassostrea virginica NCBI Genome Assembly using RepeatMasker 4.07

0000-0002-2747-368X

Genome used: NCBI GCA_002022765.4_C_virginica-3.0

I ran RepeatMasker (v4.07) with RepBase-20170127 and RMBlast 2.6.0 with species set to Crassotrea virginica.

All commands were documented in a Jupyter Notebook (GitHub):

Sam’s Notebook:Transposable Element Mapping – Olympia Oyster Genome Assembly using RepeatMasker 4.07

0000-0002-2747-368X

Steven wanted transposable elements (TEs) in the Olympia oyster genome identified.

After some minor struggles, I was able to get RepeatMasker installed on on both of our Apple Xserves (emu & roadrunner; running Ubuntu 16.04LTS).

Genome used: pbjelly_sjw_01

I ran RepeatMasker (v4.07) with RepBase-20170127 and RMBlast 2.6.0 four times:

  1. Default settings (i.e. no species select – will use human genome).
  2. Species = Crassostrea gigas (Pacific oyster)
  3. Species = Crassostrea virginica (Eastern oyster)
  4. Species = Ostrea lurida (Olympia oyster)

The idea was to get a sense of how the analyses would differ with species specifications. However, it’s likely that the only species setting that will make any difference will be Run #2 (Crassostrea gigas).

The reason I say this is that RepeatMasker has a built in tool to query which species are available in the RepBase database (e.g.):

 RepeatMasker-4.0.7/util/queryRepeatDatabase.pl -species "crassostrea virginica" -stat 

Here’s a very brief overview of what that yields:

  • Crassotrea gigas: 792 specific repeats
  • Crassostrea virginica: 4 Crassostrea virginica specific repeats
  • Ostrea lurida: 0 Ostrea lurida specific repeats

All runs were performed on roadrunner.

All commands were documented in a Jupyter Notebook (GitHub):

NOTE: RepeatMasker writes the desired output files (*.out, *.cat.gz, and *.gff) to the same directory that the genome is located in! If you conduct multiple runs with the same genome in the same directory, it will overwrite those files, as they are named using the genome assembly filename.

Grace’s Notebook: May 23, 2018, RNA isolations for warm day 12 and master data file organization

RNA isolation

Because only three crabs exposed to the warm temperature treatmeent survived the experiment, we decided during out meeting on Tuesday (2018-05-22) that I would isolate RNA from warm treatment crabs (infected and uninfected) that made it to the temperature treatment stage (day 12). So today I started that process and isolated RNA from 8 samples (4 crabs, all negative for Hematodinium based on Pam’s qPCR results).

Hemolymph collection data of samples I processed today:
img

RNA HS Qubit results from those samples (tube number ties back to FRP in the hemolymph data sheet):
img

Tomorrow I will isolate RNA from two more crabs that are warm temperature treatment and uninfected (6 total uninfected) and 6 crabs that are warm treatment infected (using qPCR data from Pam).

Data organization

Working on figuring out a way to use R to organize the data into a master file such that each row is an individual crab (as designated by the unique FRP ID number), with columns for each piece of important data we have associated with that crab (hemolymph tube numbers from the sample dates, RNA isoaltion and Qubit data, morphology data, qPCR results). I am enjoying the puzzle-solving-like aspect of this process, but it can be overwhelming sometimes becuase it is a lot of data that is in many different sheets and workbooks. Our master file will include ALL crabs, including those that died after the second sampling date.

Thursday and Friday goals:

Thursday

  • finish isolating RNA
  • data organization and master file creation

Friday

  • data organization and master file creation
  • work on and finish a pooling scheme for RNA sequencing using the master file

from Grace’s Lab Notebook https://ift.tt/2kinVv8
via IFTTT

Sam’s Notebook:DNA Received – Sea lice DNA from Cris Gallardo-Escarate at Universidad de Concepción

0000-0002-2747-368X

Received Caligus tape DNA – two samples:

  • Female 1 .
  • Female 2 .

Stored in slots H4 and H5 in “Sam’s gDNA Box #2″ in the FTR 213 -20oC freezer.

Google Sheet: Sam’s gDNA Box #2

20180523_sea_lice_dna.jpg

from Sam’s Notebook https://ift.tt/2LnZTv3
via IFTTT

Sam’s Notebook:Software Installation – RepeatMasker v4.0.7 on Emu/Roadrunner Continued

0000-0002-2747-368X

After yesterday’s difficulties getting RMblast to compile, I deleted the folder and went through the build process again.

This time it worked, but it did not put rmblastn in the specified location (/home/shared/rmblast).

This fact took me a fair amount of time to figure out. Finally, after a couple of different re-builds, I ran find to see if rmblastn existed somewhere I wasn’t looking:

20180523_rmblast_install_01.png

Additionally, I couldn’t find the location of the various BLAST executables. Some internet sleuthing led me to the NCBI page on installing BLAST+ from source, which indicates that the executables are stored in:

 ncbi-blast-VERSION+-src/c++/ReleaseMT/bin/ 

How intuitive! /s

In order to improve readability and usability of the /home/shared/ directory, I renamed the /home/shared/rmblast directory to reflect the BLAST version and created a symbolic link in that directory to the rmlbastn executable:

Symbolic link to RMBLAST

20180523_rmblast_install_02.png

Initiate RepeatMasker configuration