Tutorial 1: Running batch analyses on the Campus Cluster, cont.
***If you are in Champaign, you should be working on this assignment as a group between 9-11:20am in 3401 Siebel Center. At 11:30am, go to 2405 Siebel Center for the weekly lunch seminar --- yay free lunch! I know that Emma needs to go to class during this time.
Everyone did a great job yesterday! I am so pleased with all of your progress -- epsecially in my absence! Before moving on to the next activity, we are going to be making some modifications to the scripts that we wrote yesterday.
Ananya and Emma: If you could clone this repository (instead of forking it) that would be great! I already added your scripts from yesterday. This will make it easier for me to see what you both are doing and it will make it easier for both of you to access the latest version of this repository.
To pull the latest changes to the repository, type
cd /projects/tallis/[YourUserName]/reu2019-tutorials git pull
If you working on this tutorial, type
cd 1-campus-cluster/[YourUserName] vim b_compare_fastme_trees.pbs
At the top of this file, copy the following text
#!/bin/bash #PBS -N "tutorial-1-campus-cluster" #PBS -W group_list=tallis #PBS -q secondary #PBS -l nodes=1:ppn=12 #PBS -l walltime=00:10:00 #PBS -j oe #PBS -M [YourNetID]@illinois.edu #PBS -m be
Below these lines, write a bash script to run that uses
compare_trees.py to compare the true tree to the estimated trees. Specifically, the script should output a CSV file, called
fastme.csv, with the following header column:
MODlis the model condition (
REPLis the replicate number (
DISTis the method used to compute distances between pairs of sequences
TREEis the method used to estimate a tree from the distance matrix
NL,I1,I2,FP,FN,RFare the output of
Yesterday, you estimated four different trees on each of the fifteen dataset, so your CSV file should have 4 * 5 * 3 + 1 = 61 lines. To check the number of lines in your CSV file, type
wc -l fastme.csv
Suppose that FastME failed on the first replicate dataset for the first model condition when you ran it uses p-distances and Neighbor Joining. Then your CSV file should include the following row:
To see a list of failures, type
grep "NA$" fastme.csv
When you finish writing your script, submit it as a job to the Campus Cluster queue; type
To see that your job has been submitted, type
qstat -u [YourNetID]
To add your script to the repository, type
git add b_compare_fastme_trees.pbs git commit -m "Add a message here" git push
You will be asked to enter your Github user name and password.