Mercurial > repos > iuc > kma

comparison test-data/generate_test_data.sh @ 0:e74a31040224 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/kma commit 43bbde4f8f8671284b2acb21dfd2657de4ba967f"
author iuc
date Sat, 15 Feb 2020 20:32:33 +0000
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:e74a31040224
1 #!/usr/bin/bash
2
3 # E. coli locus b0842 (b0842.fasta.gz) downloaded from Enterobase E. coli cgMLST scheme
4 # requires: wget, kma, bwa, samtools, bedtools
5
6 gunzip b0842.fasta.gz
7
8 # Take first 5 alleles to reduce size of test data
9 mkdir ecoli_cgMLST
10 head -n 10 b0842.fasta > ecoli_cgMLST/ecoli_b0842_1to5.fasta
11
12 kma index -k 8 -i ecoli_cgMLST/ecoli_b0842_1to5.fasta -o ecoli_cgMLST/ecoli_b0842_1to5
13
14 wget ftp://ftp.sra.ebi.ac.uk/vol1/fastq/ERR884/ERR884056/ERR884056_1.fastq.gz
15
16 # Use bwa to map reads to reduced E. coli locus b0842
17 # and extract only mapped reads (to reduce size of test dataset)
18 bwa index ecoli_cgMLST/ecoli_b0842_1to5.fasta
19
20 bwa mem ecoli_cgMLST/ecoli_b0842_1to5.fasta ERR884056_1.fastq.gz -o ERR884056_1_ecoli_b0842_1to5.sam
21
22 samtools view ERR884056_1_ecoli_b0842_1to5.sam -bo ERR884056_1_ecoli_b0842_1to5.bam
23
24 # Select mapped reads
25 samtools view -b -F 4 ERR884056_1_ecoli_b0842_1to5.bam > ERR884056_1_ecoli_b0842_1to5.mapped.bam
26
27 samtools sort -n ERR884056_1_ecoli_b0842_1to5.mapped.bam -o ERR884056_1_ecoli_b0842_1to5.mapped.sort.bam
28
29 bedtools bamtofastq -i ERR884056_1_ecoli_b0842_1to5.mapped.sort.bam -fq ERR884056_ecoli_b0842.mapped_R1.fastq
30