[khmer] khmer updates: the bleeding-edge branch.

C. Titus Brown ctb at msu.edu
Tue Aug 20 09:31:34 PDT 2013

Hi all,

we're making good progress towards something that looks like a major revamping of khmer.  This next release of khmer will feature significant improvements in speed, testing, and documentation.  We're also producing some standardized protocols for doing mrnaseq assembly, metagenome assembly, and (soon) genome assembly with khmer.

**If you are happily using the default, master branch of khmer, nothing has changed and you don't need to read this e-mail.*  But if you're feeling adventuresome, read on :).

I'm sending this e-mail in part because I've just broken a bunch of my recent tutorials in the process of making various scripts more stable, better documented, etc.  Specifically, the three scripts


which are staples of read manipulation for khmer have now been renamed to


As part of this they have been placed under test and documented. (!!)

Of critical note, both filter-abund and normalize-by-median have newly expanded behavior; filter-abund now takes a '-V' option for use on variable coverage data like metagenomes and transcriptomes,, and normalize-by-median both has a paired-end option (courtesy of Jared Simpson - see -p) and a new robustness to errors (courtesy of Chris Welcher).


You can read the new expanded documentation here:


especially for the scripts,


We are also soon to announce some standardized protocols for doing mrnaseq and metagenome assembly.  If you want to take a look, the mRNAseq protocol is under development here:


and the metagenome assembly protocol is being adapted from this tutorial: http://ged.msu.edu/angus/2013-stamps/

If you have feedback on the protocol please send it to me privately, for now; we'll be discussing protocols on a new list, which I haven't announced yet.

I'll send the various announcements to this list, so stay tuned.


More information about the khmer mailing list