write human screening tutorial that derives from the previous SRA submission code about qiime HOT 4 OPEN

Imported from trac issue 14. Created by caporaso on 2011-04-25T14:12:12, last modified: 2012-08-31T16:42:30

from qiime.

Trac comment by caporaso on 2012-08-31 16:38:09:

Discussed this with Dirk Gevers -- see this information from him:

I just came across the commandline we've been using here:

bmtagger.sh -q 0 -1 rep_set.fna -b humanGenome/dna.fna.bitmask -x humanGenome/dna.fna.srprism -T . -o human-like.reads.list

you'll find bmtagger here - ftp://ftp.ncbi.nlm.nih.gov/pub/agarwala/bmtagger/ (publication is in the works)

and the Human Genome DBs are being placed here:
ftp://longrna:[email protected]//MiSeq/humangenome.tgz (note that the tar-cmd is currently still running, and should likely be completed tomorrow)

I'll get the info on how we got to this and the human genome DB to you once both Kat and me are back in the office (9/17)

from qiime.

Trac comment by caporaso on 2012-08-31 16:42:30:

Note that if the above ftp link breaks, this data is on compy at /scratch/caporaso/humangenome.

from qiime.

At the same time, we should re-factor the exclude_seqs_by_blast.py script to use other mapping tools as the BLAST-based methods likely won't scale to modern-sized data sets, and a global-alignment-based approach (e.g., usearch is likely more relevant).

from qiime.