cleanup

cre.bcbio.upgrade.sh

@@ -23,17 +23,18 @@ date
 # create a .test_profile:
 # export PATH=$HOME/cre:$HOME/crt:$HOME/crg:/hpf/largeprojects/ccmbio/naumenko/tools/bcbio_1.1.5/anaconda/bin:$HOME/tools/mc-4.8.16/bin:$HOME/jkent_tools:$HOME/bioscripts:.:/usr/local/bin:/opt/moab/bin:/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/sbin
 # export PYTHONPATH=/hpf/largeprojects/ccmbio/naumenko/tools/bcbio_1.1.5/anaconda/lib/python3.6
-. /hpf/largeprojects/ccmbio/naumenko/tools/bcbio_1.1.5/.test_profile
+. /hpf/largeprojects/ccmbio/naumenko/tools/bcbio_1.1.5/.profile115
 which python
 echo $PYTHONPATH
 ######################################################################
 # 3. Upgrade tools. If tooldir was set before, no need to specify it again
 which bcbio_nextgen.py
-# bcbio_nextgen.py upgrade -u skip --tools 
+bcbio_nextgen.py upgrade -u skip --tools
 #--tooldir $1
 ######################################################################
 # 4. Install indices
-# bcbio_nextgen.py upgrade -u skip --genomes GRCh37 --aligners bwa --cores 10
+# genomes = {GRCh37, hg38}
+# bcbio_nextgen.py upgrade -u skip --genomes hg38 --aligners bwa --cores 10
 # bcbio_nextgen.py upgrade -u skip --genomes GRCh37 --aligners star --cores 10
 # bcbio_nextgen.py upgrade -u skip --genomes GRCh37 --aligners hisat2 --cores 10
 # bcbio_nextgen.py upgrade -u skip --genomes GRCh37 --aligners rtg --cores 10
@@ -49,7 +50,7 @@ which bcbio_nextgen.py
 # 1.1.5 - a huge update to python3, installed from scratch
 
 # upgrades data installed before (gemini, cadd) for all references
-# bcbio_nextgen.py upgrade --data
+bcbio_nextgen.py upgrade --data --genomes GRCh37
 
 # VEP is upgraded quite often ~2-3 months - when upgrading tools it looks for new VEP cache
 # bcbio_nextgen.py upgrade -u skip --genomes GRCh37 --datatarget vep

cre.bcbio.upgrade.star.sh

@@ -1,19 +1,22 @@
 #!/bin/bash
-#PBS -l walltime=240:00:00,nodes=1:ppn=30
+#PBS -l walltime=150:00:00,nodes=1:ppn=30
 #PBS -joe .
 #PBS -d .
-#PBS -l vmem=100g,mem=100g
+#PBS -l vmem=50g,mem=50g
 
-. /hpf/largeprojects/ccmbio/naumenko/tools/bcbio_1.1.5/.test_profile
+. /hpf/largeprojects/ccmbio/naumenko/tools/bcbio_1.1.5/.profile115
 which bcbio_nextgen.py
 
 hostname
 
 bcbio_path=/hpf/largeprojects/ccmbio/naumenko/tools/bcbio_1.1.5
 
+#GRCh37
+reference="hg38"
+
 # sometimes cannot upgrade STAR on data nodes - memory is low, or cannot take much CPUs
 # can take more than a day
 export PATH=${bcbio_path}/bin:$PATH && STAR \
---genomeDir ${bcbio_path}/genomes/Hsapiens/GRCh37/star \
---genomeFastaFiles ${bcbio_path}/genomes/Hsapiens/GRCh37/seq/GRCh37.fa \
+--genomeDir ${bcbio_path}/genomes/Hsapiens/${reference}/star \
+--genomeFastaFiles ${bcbio_path}/genomes/Hsapiens/${reference}/seq/${reference}.fa \
 --runThreadN 30 --limitGenomeGenerateRAM 30000000000 --genomeChrBinNbits 14 --runMode genomeGenerate --genomeSAindexNbases 14

cre.bcbio.upgrade.star.sh.o47757114

@@ -0,0 +1,14 @@
+/hpf/largeprojects/ccmbio/naumenko/tools/bcbio_1.1.5/bin/bcbio_nextgen.py
+node080
+May 25 22:55:29 ..... started STAR run
+May 25 22:55:30 ... starting to generate Genome files
+May 25 22:56:48 ... starting to sort Suffix Array. This may take a long time...
+May 25 22:57:14 ... sorting Suffix Array chunks and saving them to disk...
+May 25 23:22:50 ... loading chunks from disk, packing SA...
+May 25 23:30:16 ... finished generating suffix array
+May 25 23:30:16 ... generating Suffix Array index
+May 25 23:35:50 ... completed Suffix Array index
+May 25 23:35:50 ... writing Genome to disk ...
+May 25 23:36:04 ... writing Suffix Array to disk ...
+May 25 23:39:16 ... writing SAindex to disk
+May 25 23:39:41 ..... finished successfully

cre.gemini2txt.vcf2db.sh

@@ -4,7 +4,7 @@
 #  when using v.chr = g.chr AND v.gene = g.gene it becomes very slow
 #  by default bcbio writes PASS only variants to the database
 
-#  example call: cre.gemini2txt.sh S28-ensemble.db 5 ALL
+#  example call: cre.gemini2txt.sh S28-ensemble.db 5 ALL 0.01
 #  when using vcfanno/vcfdb loader some fields are different
 #  for some reason \n in the query string does not work here