jhlr/indelminer

Identify indels from BAM files.

indelMINER

Identify indels from BAM files.

REQUIREMENTS

indelMINER should work on any standard 64 bit Linux environment with

• GCC
• zlib (http://zlib.net)

ACKNOWLEDGEMENTS

indelMINER uses Paul Hsieh's Superfasthash (http://www.azillionmonkeys.com/qed/hash.html) hash function. The hash function is encapsulated in the source file superfasthash.c within this distribution. indelMINER also includes the 0.1.19 version of SAMtools by Heng Li as part of the distribution.

I would like to thank the following users for bug reports and other contributions towards the on-going development of indelMINER.

• Trent Walradt, Yale School of Medicine
• Pieter Busschaert, University of Leuven

SUMMARY

indelMINER refers to a set of algorithms to identify indels from whole genome resequencing datasets using paired-end reads. indelMINER uses a split-read approach to identify the precise breakpoints for indels of size less than a user specified threshold, and supplements that with a paired-end approach to identify larger variants that are frequently missed with the split-read approach.

INSTALLATION

indelMINER uses the SAMtools (http://samtools.sourceforge.net) API to manipulate files in the BAM (Binary Alignment/Mapping) format. Current version of indelMINER comes with the version 0.1.19 of SAMtools included as part of the distribution.

Type:

make 

This should create an binary named indelminer in the "src" folder. Add the "src" folder to your PATH or use the full path to the binary when running indelMINER

TEST-DATASET

The test_data directory contains the following files

• reference.fa : A reference sequence in FASTA format.
• alignments.bam : An alignment file in the BAM format.
• alignments.bai : Index for the provided alignment.bam.
• indelminer.config : An optional file that can be used to speed up a run.
• indelminer.expected.vcf : A list of expected variants in the BAM file in VCF format.

The alignment file in the BAM format for 100x100 bp paired end sequences with an mean insert length of 500 bps. The sequences were aligned to the reference sequence in reference.fa using BWA, then cleaned using GATK IndelRealigner. The indel alignments.bai was generated on a x86_64 system, so please reindex if you are attempting to run on a different platform.

To run indelMINER on the provided BAM file, do the following (in the test_data folder)

./../src/indelminer reference.fa sample=alignments.bam > indelminer.flt.vcf

The expected output is provided in the file "indelminer.expected.vcf". You can do a simple "diff" to compare the output in "indelminer.expected.vcf" to the output generated by your run in "indelminer.flt.vcf".

The first step in indelMINER involves calculation of the minimum and maximum proper insert lengths for all the read-groups in the BAM, and the expected average coverage on all the contigs/chromosomes in the reference sequence. The user can speed up indelMINER by providing these information in a configuration file to indelMINER. The configuration file should contain lines in the following format:

IL read_group min_insert max_insert

to specify that a read group (read_group) contains pairs with the minimum and maximum proper insert lengths of min_insert and max_insert.

Similarly the chromosomal coverage can be specified using the following format:

RC chromosome_name average_coverage

There is a file called indelminer.config in your test_data directory which specifies these values for the BAM file included with the distribution. In order to use it in a run of indelMINER, please do the following:

./../src/indelminer -i indelminer.config reference.fa sample=alignments.bam > indelminer.flt.vcf

APPLICATION NOTES

TUMOR/NORMAL PAIRS

In cases where you have a BAM file of alignments from a tumor sample, the first step is to run indelMINER on the tumor sample.

indelminer reference.fa tumor=tumor.bam > tumor.vcf

This identifies all the indels in the tumor sample. You might want to change a few of the arguments to indelMINER in this command based on yoru dataset. Specially you might want to supply a configuration file, which might make indelMINER run faster. You might want to change "-b" if you have smaller size reads (The defaults work well for 100 bp reads).

One you have the indels in the tumor, run indelMINER to annotate those indels based on their presence/absence in the normal.

indelminer -q 0 -a -e 1 reference.fa tumor.vcf normal=normal.bam > tumor.normal.vcf

This adds an annotation "normal" to each variant in tumor.vcf if it is also seen in the normal. Using "-q 0 -a -e 1" ensures that any sign of its presence in the tumor is considered.