ncbi-codeathons / bothie Goto Github PK

ref https://github.com/ekimb/rust-mdbg and https://www.biorxiv.org/content/10.1101/2021.06.09.447586v1

I think this is maybe a job for @rchikhi :)

per https://github.com/dkoslicki/CMash, we can preprocess the sample with MakeNodeGraph.py during the sample preparation step.

as a tool, would be good to show these results too

slides

some bothie results - spreadsheet

In #17, @bluegenes noted that in our timelog output, we do not indicate which tool is being run.

my hot take is that it's better to have all the tool-specific things under one common directory.

• support tool and config file columns in the CSV
• support a 'ground truth' column in the CSV

hi team, over in #32, I ran metagraph like so:

./run-many-csv metagraph/conf-basic.yml metagraph data/data-small.csv -o out.csv

and the resulting out.csv included this line:

data/genome-s10+s11.fa.gz,data/random-query.fa,1,outputs.metagraph.8t1f9hu3/results.1.3

show that it found the negative control I added.

Looking at all the results files together with

find outputs.metagraph.8t1f9hu3 -name results.csv -exec cat {} \;

I see:

0       random10k
0       s10.4000        /disk/bothie/data/genome-s10+s11.fa.gz
0       s10.4000        /disk/bothie/data/genome-s10.fa.gz
0       s10.100 /disk/bothie/data/genome-s10+s11.fa.gz
0       s10.100 /disk/bothie/data/genome-s10.fa.gz

so it looks like metagraph isn't properly working yet, maybe.

cc @jackroddy @phiweger

TODO!

https://en.wikipedia.org/wiki/Bothy

cc @laitanawe

Slack is down for some users including me, so until it's up again, I'll convert this to chat.

@jackroddy @ctb How should we move metagraph forward?