Notes to ourselves

Notes to ourselves

Tips, tricks and gotchas for our future selves.

Reading BCFs in IGV

IGV will not take the bcf, and it will only take a vcf if it is bgzip compressed and if the file ending is vcf.gz and has a tabix index.

file=freebayes.bcf
bcftools view -Ov ${file} > ${file%%.bcf}.vcf
bgzip ${file%%.bcf}.vcf
tabix ${file%%.bcf}.vcf.gz

See this for more info! See below if you’re not sure what’s happening in the curly braces.

Sorting a multi-fasta file by name

This is useful for when a reference genome does not come with chromosomes or contigs in a sensible order. The sorting order could be changed by modifying the sort -n below to suit.

FASTA_TO_SORT=somefile.fasta

# Check if the fasta file has been indexed already, if not, index it
test -f ${FASTA_TO_SORT}.fai || samtools faidx ${FASTA_TO_SORT})

# Grab out the sequences by name in the correct order.
samtools faidx ${FASTA_TO_SORT} \
    $(cut -f1 ${FASTA_TO_SORT}.fai | sort -n) \
    > file_sorted.fasta

The beauty of this is that it uses very little RAM even on huge references, as it uses samtools faidx to perform random access into the fasta file.

Bash variable magic

Remove extension:

fn=test.ext
bn=${fn%%.ext}
echo "basename of $fn is $bn"

The TLDP site has as bunch of great resources for this

Raijin Rmpi

You need to install it with:

install.packages("Rmpi", configure.args=c("--with-Rmpi-include=/apps/openmpi/1.6.3/include",
                                          "--with-Rmpi-libpath=/apps/openmpi/1.6.3/lib",
                                          "--with-Rmpi-type=OPENMPI"))