gatoravi

#List all the instrument-data that is imported
genome instrument-data list imported "sample.patient.common_name='TST1ds' and sample.extraction_type = 'genomic dna'"
genome instrument-data list imported "sample.patient.common_name='TST1ds' and sample.extraction_type = 'rna'"

#Define the models based on imported instrument-data, rename the ID's
genome model define rna-seq  --reference-sequence-build='106942997'  --annotation-build='124434505'  --cancer-annotation-db='tgi/cancer-annotation/human/build37-20131010.1' --subject='H_NJ-HCC1395ds-HCC1395_RNA'  --processing-profile='2762841'  --instrument-data='0296e0aeeb3f4daab1e8604cddb2cb70' --model-name='hcc1395ds-tumor-rnaseq'
genome model define rna-seq  --reference-sequence-build='106942997'  --annotation-build='124434505'  --cancer-annotation-db='tgi/cancer-annotation/human/build37-20131010.1' --subject='H_NJ-HCC1395ds-HCC1395_BL_RNA'  --processing-profile='2762841'  --instrument-data='675561c466744c8498f903b77ee9acbe' --model-name='hcc1395ds-normal-rnaseq'
ge

gatoravi

#! /bin/Rscript

run_em <- function(data, cutoff = 1.50, theta_init = 0.5, iter = 100, tolerance = 1e-4) {
    missing = data[data == cutoff]
    observed = data[data != cutoff]
    n1 = length(observed)
    n2 = length(missing)
    n = n1 + n2
    theta_current = theta_init
    theta_new = 0

gatoravi

ls *tsv | xargs -i echo 'cut -f 1-4,6-9 {} |tr ":" "\t" | cut -f 1-4,6,20,34,48 | perl -ne '\''chomp$_; @array=split("\t"); if($array[3] == $array[4] || $array[3] == $array[5] || $array[3] == $array[6] || $array[3] == $array[7]){next;} else {print "$_\n";}'\'' > {}.formatted '

gatoravi

#VEP is on clia1 here - /opt/gms/vep/variant_effect_predictor

#An example that I used to run it was,
#VEP is installed here /opt/gms/vep/variant_effect_predictor
#PERL5LIB=$PERL5LIB:/opt/gms/vep/ perl variant_effect_predictor.pl -i example_GRCh37.vcf --cache --dir_cache /opt/gms/vep/

#another example with required VEP annotations for GEMINI
PERL5LIB=$PERL5LIB:/opt/gms/vep/ perl /opt/gms/vep/variant_effect_predictor/variant_effect_predictor.pl -i test.vcf \
--cache --sift b --polyphen b --symbol --numbers --biotype --total_length \
-o snvs.detailed.vep.vcf --vcf --fields Consequence,Codons,Amino_acids,Gene,SYMBOL,Feature,EXON,PolyPhen,SIFT,Protein_position,BIOTYPE \

gatoravi

#! /bin/bash

#This script takes 2 arguments
#The first script is the BAM file
#The second is the BED file

bam=$1
bed=$2
awk -v bam=$bam 'BEGIN { print "new\ngenome hg19\nload "bam"\nsnapshotDirectory ./snapshots/" } { gsub("chr", "", $1); if(NR > 1) print "goto "$1"\t"$2-10"\t"$2+10"\nsort base\ncollapse\nsnapshot"; }' $bed

gatoravi

# Compiled source #
###################
*.com
*.class
*.dll
*.exe
*.o
*.so

# Packages #

gatoravi

1. The Practice of programming - Brian W. Kernighan
2. The Unix Programming Environment - Brian W. Kernighan and Rob Pike
3. Numerical Recipes in C: The Art of Scientific Computing
4. Structure and Interpretation of Computer Programs.
5. Higher-Order Perl: Transforming Programs with Programs - Mark Jason Dominus
6. Reverse Engineering for Beginners - Dennis Yurichev
7. The Pragmatic Programmer: From Journeyman to Master
8. The Mythical Man-Month

gatoravi

#! /usr/bin/Rscript

test_read_write <- function() {
    iris_copy <- read.table("iris_copy.tsv", head = T, sep = "\t")
    iris_copy$Sepal.Length <- iris$Sepal.Length + 1
    write.table(iris_copy, "iris_copy.tsv", row.names = F, quote = F,
                sep = "\t")
}

create_original_modified_iris <- function() {

gatoravi

##Exiting from programs Exceptions aren't the only way. There are times when cerr/exit are acceptable, but personally, I almost never do that. The answer to your question is kind of long.

I tend to think like a library developer (i.e., code reuse is always on my mind). If your function calls exit(), that makes it impossible to use in other contexts that might not care about whatever error condition is causing you to want to exit. "But what about contexts where the error is truly unrecoverable?" In those cases, exit() inhibits debugging, use abort() instead. This should be reserved for truly egregious errors--those which you'll want to get a core dump

gatoravi

#Generate the FASTA
rm -f test.fa; samtools faidx ../../regtools-test/chr22.fa 22:41475000-41585000 > test.fa
sed "1s/.*/>22/" test.fa > new_test.fa; rm -f test.fa; mv new_test.fa test.fa #fix the header
samtools faidx test.fa

#Generate the GTF
rm -f test.gtf; awk -F"\t" '$4> 41480000 && $5 < 41580000 { OFS = "\t"; $4 = $4 - 41475000 + 1; $5 = $5 - 41475000 + 1; print }'  ../../regtools-test/genes_chr22.gtf > test.gtf

#Create the junctions.bed
rm -f test_junctions.bed; awk '$1==22 && $2> 41480000 && $3 < 41580000 { OFS = "\t"; $2=$2 - 41475000 + 1; $3 = $3 - 41475000 + 1; $7=$2; $8=$3; print }'  hcc1395_junctions.bed > test_junctions.bed