Skip to content

Instantly share code, notes, and snippets.

@erikenbody
Last active May 26, 2016 18:39
Show Gist options
  • Save erikenbody/3ff782f5fa43d2f2802aa0f2eaa2572b to your computer and use it in GitHub Desktop.
Save erikenbody/3ff782f5fa43d2f2802aa0f2eaa2572b to your computer and use it in GitHub Desktop.

##Notes May 18th Meeting

##Dependencies

-For context, I am running OS X El Capitan 10.11.14 on a MacBook Pro. Everything is running in bash on the OSX terminal.

-run pip to install anything python related. Everything else related to pyRAD, including pyRAD, was downloaded from their respective git repositories.

pip install scipy 
pip install numpy

-These were great for installing muscle and vsearch using brew. However, vsearch is outdated.

brew install homebrew/science/muscle
brew install homebrew/science/vsearch

-Following the vsearch website, I downloaded the binary of vsearch for osx using these commands.

wget https://github.com/torognes/vsearch/releases/download/v1.11.1/vsearch-1.11.1-linux-x86_64.tar.gz
tar xzf vsearch-1.11.1-linux-x86_64.tar.gz

-Following massive headaches and exploration and many dead ends trying to get this version in my path, here was my solution for running vsearch: Put the whole vsearch folder in my home folder - then linked to this in the params file later (it would not work anywhere but my home folder!!!) - so line four in the params file looks like this for me:

~/vsearch-1.11.1-osx-x86_64/bin/vsearch ##4. full path

-UPDATE I was able to call it from anywhere by setting a path to this home folder. Maybe my original issue was involved with spaces and backslashes associated with google drive?

export PATH=~/vsearch-1.11.1-osx-x86_64/bin/:$PATH

-had to update pip and brew and all these things. I also had to update python

-Graham did a bunch of things that were quite confusing to get my pyRAD.py into my python path.End of the day put the following two things in my bash profile. Edit .bashrc for a profile that you have to source every time you open the terminal.

nano .bashrc

export PYTHONPATH='usr/local/lib/python2.7/site-packages:/Users/erikenbody/Google Drive/Tulane/Conferences_and_Workshops/Tulane_Summer_Genomics_Workgroup/pyrad/pyrad'
export PATH="$PATH:/Users/erikenbody/Google Drive/Tulane/Conferences_and_Workshops/Tulane_Summer_Genomics_Workgroup/pyrad/pyrad">'

source .bashrc

###pyRAD RAD tutorial

####Running Step 1

-I'll work through this tutorial and only make notes where I had issues beyond what was obviously stated already.

-In RAD tutorial ln 4 these commands did not work so Graham edited them. However, params file didn't come out clean so there are still issues.

sed -i -e '/## 7. /c\'$'\012''2                   ## 7. N processors... ' params.txt
sed -i -e '/## 10. /c\'$'\012''.85                ## 10. lowered clust thresh... ' params.txt
sed -i -e '/## 14. /c\'$'\012''c85m4p3            ## 14. outprefix... ' params.txt
sed -i -e '/## 24./c\'$'\012''8                   ## 24. maxH raised ... ' params.txt
ed -i -e '/## 30./c\'$'\012''*                   ## 30. all output formats... ' params.txt

-he also suggested this. Which still didn't quite work.

sed -i -e '/$$## 7. / s/&^/../2/' params.txt

-For now I will just edit params.txt

nano params.txt

-I also tried just editing it using my favorite text editor and this worked fine too. It also worked to copy the text from the tutorial and paste it into a new params.txt file. Its worth familiarizing with what is in this file via the General tutorial. For example, it might be nice to set the working directory for the current project this way.

####May 17th notes (working on my own) -I had to locate the .bashrc profile which was in my home directory. Considering adding this to my .bash_profile so it is always added at login. Downsides to that?

After sourcing my new .bash_profile that contains the path information, I was able to run pyRAD as such

pyRAD.py

-However, it said it could not find numpy and scipy again! I had to go reinstall these with pip. Why did this happen?

pip install scipy
pip install numpy

-random useful command. This shows the environment you're running in, including PATH and PYTHON path. Also terminal profile information.

env

For inspecting the fastq file, I had to unzip the RAD file. Then view with less like in the tutorial. Alternatively, you can replace less with zless for this and the future commands.

gunzip simRADs_R1.fastq

Here is information on .fastq files by line:

The fastq file Each read has 4 lines of information, which include the following: Sequence identifier Sequence Quality score identifier line (consisting of a +) Quality score Each sequence identifier, the line that precedes the sequence and describes it, needs to be in the following format: @:::::: :: :

  • @ @ Each sequence identifier line starts with @
  • Characters allowed: Instrument ID a-z, A-Z, 0-9 and underscore
  • Numerical Run number on instrument
  • Characters allowed: a-z, A-Z, 0-9
  • Numerical Lane number
  • Numerical Tile number
  • <x_pos> Numerical X coordinate of cluster
  • <y_pos> Numerical Y coordinate of cluster
  • Numerical Read number. 1 can be single read or read 2 of paired-end
  • Y or N Y if the read is filtered, N otherwise
  • Numerical 0 when none of the control bits are on, otherwise it is an eve

-For line 9, I hit errors while running pyRAD. It turns out it was because my working folder was Goolge Drive and has a space in it. I've since edited it to be Google_Drive, but alternatively you could run this out of your home folder. No issues running step 1.

##Steps 2-7

-keep in mind running on unix on my mac I have to run pyRAD as pyRAD.py. -This was pretty cookie cutter with no problems through step 7

pyRAD.py -p params.txt -s 3
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment