Closed steven-albanese closed 8 years ago
Oh- I also commented out a few lines where the script was trying to pickle some generators, which was giving me an error message as well. They don't appear to be used elsewhere at the moment, but I can revert those as well if wanted
Thanks for finding and correcting these issues!
Looks good to me.
@maxentile and I noticed a problem where using a path_to_trajs
with a wild card would still return only a single trajectory when printing fnames
. Seems like using single quotes around the trajectory fixes that problem
Awesome! Are there any sample plots being generated with this to show what the pipeline is doing so far? If so, I'd love to see them! Maybe share them on the #msm slack channel?
sorry @jchodera, I just saw your comment. The plots being generated here are the same as I showed you when we met earlier today. Maybe we could run the pipeline on a more straightforward case than mTOR to see? I've got it chugging away on an EGFR project over the weekend
Ok this should be ready to merge, I think I've cleaned everything up appropriately!
Thanks for fixing these typos, Steve. I hadn't been able to get the image making scripts to work, yet, so this is great! And yeah the pickling problem is already in issue #2. Will merge! (We can add show plots in a later PR or on Slack.)
So, I found a bug where
project_name
wasn't being passed toplot_sanity_check
, which was throwing up an error. Just wanted to merge in the fix for that. I also addedpipeline-torque.sh
, which is a shell script for running the pipeline on the batch queue on the cluster.There are some other small changes, like
max_iter
, which I doubled because 500 steps wasn't enough to converge clustering the WT mTOR kinase domain data, but I don't think that's necessarily a change that needs to be included here. I can revert it, if you'd like