Closed seb-mueller closed 5 years ago
Agreed! Let's get everything in segmentation_2018. The annotation will need re-running anyway so the only file which is in its final version at the moment is the segmentation.
Brilliant! Hope you talk went smooth! One last thing, I was trying to move things around but some of the files are locked, could you run this quickly giving me access? I've done the same already:
chmod -R g+w /projects/nick_matthews
Hi Seb, should now be done, let me know if any issues!
Update: I've cleaned up quite a bit, pleas let me know if I overdid it (details upon inspecting the commits). Effectively, I've chucked out all non used or obsolete files into archive
, also from the base directory. I'm still not finished, but just wanted to let you know in case you want to have a look.
Heyyy, had a quick look and look generally sensible.
Re: counting biases are the objects required for repetativeness class actually in the local environment when running the annotation pipeline?: https://github.com/seb-mueller/chlamy_locus_map/blob/8a754a16638635c1ac01ea3d528b03e7157bb8d3/Scripts/chlamy_annotation_pipeline.R#L167
There's also a new empty function called processing biases in the source code, I'm assuming this is a working progress?
Just seen comment on other thread, let me know when you finished playing around with the counting biases and I'll give it a check through.
Haha, you're ahead of time :) Yes, this line is gonna change (the rowSums are being precomputed in the gr
object). I found it more sensible to have all parameter and path settings in one document..
Pretty happy with the clean-up. One more thing might to put the 3 scripts from Key-coding-scripts into scripts the reduce folders, any objection? If not, I'll go ahead and close this issue.
All looks good to me! Key-coding-scripts was just the originals from back in 2016, we've pretty heavily adapted these now, and I have copies on my computer.
I've noticed there are multipe output directories, i.e.
segmentation_2018
,phasing
andPlots
within the github. Having many other projects, I'd like to streamline (so it's easy to understand instantly), could we agree on one output location if you don't object?Since plots and other output tend to be large, I'd suggest to save outside the git repo to not clog it up to much, could we save everything in
segmentation_2018
with some sort of version indication?I could try to do this since I'd like to go through the code and clean/check it anyway making this sort of adaption along the way. Before I start I'd like to make sure you haven't go any changes left to add so I can start from the newest version?