Permalink
Browse files

Use nanopolish extract to build fasta files on 1d libraries.

  • Loading branch information...
barneypotter24 committed Sep 8, 2017
1 parent 981e24d commit fa53f0a7909eb7e1cd0926eb1d65503123f27cea
Showing with 2 additions and 2 deletions.
  1. +2 −2 pipeline/README.md
View
@@ -52,9 +52,9 @@ Albacore writes basecalled `fast5` files into subdirectories within the `workspa
If this is the case, then you will need to split the files back up into smaller subdirectories, and you'll need to submit a Porechop job _for each subdirectory_. If your library has more than 4 million reads, you can use the [`split_1d_library.py`](demux/split_1d_library.py) to sort reads into directories that contain 500,000 reads per directory.
##### Extract fasta files from basecalled reads and demultiplex reads based on barcoding:
1. Load Poretools with `module load Python/2.7.13-foss-2016b-fh2` (Uses Python 2).
1. Load Poretools with `ml nanopolish/0.7.1-foss-2016b` (Uses Python 2).
2. Change working directory to `<BASECALLED_READS_DIRECTORY>/workspace/demux/`.
3. Submit job as `sbatch --time=48:00:00 --mem=20000 --mail-type=END,FAIL --mail-user=<EMAIL_ADDRESS> --wrap="poretools fasta ../ > <FILENAME.fasta>`.
3. Submit job as `sbatch --time=96:00:00 --mem=32000 --mail-type=END,FAIL --mail-user=<EMAIL_ADDRESS.org --wrap="$EBROOTNANOPOLISH/nanopolish extract -b albacore -t template -o nanopolish_full.fasta ../"`.
4. Load Porechop with `module load Python/3.5.2-foss-2016b-fh1` (uses Python 3).
5. Submit job as `sbatch --time=24:00:00 --mem=20000 --mail-type=END,FAIL --mail-user=<EMAIL_ADDRESS> --wrap="porechop -i <FILENAME.fasta> -b . --barcode_threshold 75 --threads 16 --check_reads 100000"`.
6. Once job completes, run `gunzip NB*` from within the directory to unzip the files in preparation for consensus genome generation.

0 comments on commit fa53f0a

Please sign in to comment.