diff preprocessing-wrapper.sh @ 0:d4690e65afcd draft

Uploaded
author bcclaywell
date Thu, 26 Feb 2015 18:16:36 -0500
parents
children 221a97bf686d
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/preprocessing-wrapper.sh	Thu Feb 26 18:16:36 2015 -0500
@@ -0,0 +1,20 @@
+#!/bin/bash
+
+source $1
+
+deduplicate_sequences.py \
+    --split-map ${SPLIT_MAP} \
+    --deduplicated-sequences-file ${DEDUP_INFO} \
+    ${INPUT_SEQS} \
+    ${DEDUP_SEQS}
+
+# adapted from yapp/bin/refpkg_align
+ref_sto=$(taxit rp ${REFPKG} aln_sto)
+profile=$(taxit rp ${REFPKG} profile)
+
+sto=$(mktemp -u).sto
+
+cmalign --cpu ${GALAXY_SLOTS:-4} -o "$sto" --sfile "${ALIGNED_SCORES}" --noprob --dnaout "$profile" "${DEDUP_SEQS}" | grep -E '^#'
+
+esl-alimerge --dna --outformat afa "$ref_sto" "$sto" | \
+    seqmagick convert --output-format fasta --dash-gap - "${ALIGNED_SEQS}"