[khmer] Duration of do-partition.py (very long !) (Alexis Groppi)

[C. Titus Brown]

On Thu, Mar 21, 2013 at 04:51:30PM +0100, Alexis Groppi wrote:
> Sorry for bothering you, but it's not clear for me :
>
> For removing the artefacts :
> Should I apply find-knots on my file.below ? (after  
> normalize-by-median.py, load-into-counting.py and filter-below-abund.py)
> Then filter-stoptags ?
> And then will I have data ready for assembly or should I perform  
> do-partition.py ? (on these artefact free data)

Good questions -- it's hard to figure out how to convey this stuff
when we don't entirely understand what's going on in the first place
with the artifacts (see

http://arxiv.org/abs/1212.0159

for our best guess).

Start with your data after filter-below-abund.  Then apply

---

python scripts/load-graph.py -x 8e9 lump corn-50m.lump.fa

python scripts/make-initial-stoptags.py lump

python scripts/partition-graph.py --stoptags lump.stoptags lump

python scripts/find-knots.py -x 2e8 -N 4 lump

python scripts/filter-stoptags.py *.stoptags corn-50m.lump.fa

---

and follow with applying straight-up partitioning to the stopfilt
file.

best,
--titus