While methods for detecting both SNPs and large scale "structural" variation from
high-throughput sequencing data have been developed, detection of medium-size (10-50bp) indel variants remains a challenge. The variants are too big to be found with mapping of short reads,
and too small to be detected with previous methods for locating structural variations.
In this paper we present MoDIL, or Mixture of Distributions Indel Locator, a novel method for finding
medium sized indels from high throughput sequencing datasets. Our method can take advantage of the high
clone coverage of these datasets to identify progressively shorter indel variants, even if the individual
clone sizes are unreliable. Furthermore, our method models each genomic locus as a mixture of two
distributions of indel sizes, allowing for the detection of heterozygous, as well as homozygous variants.
Jan 20, 2012 -
The source code is cleaned up and the speed is improved.
July 6, 2010 -
Unnecessary packages are cleared in the source code.
(It could make problems for installation.)
November 28, 2009 -
There was a problem in running MoDIL on user's own dataset.
MoDIL_simple.py is fixed.
August 31, 2009 -
A bug in setup.py is fixed.
August 30, 2009 -
Source code (beta version) was uploaded to the website.
April 22, 2009 -
Version 1.1 of results were uploaded to the website.