Difficulty mapping reads with non-reference allele?

[krobison]

I thought I recently read a paper detailing poorer read mapping to a reference genome if the read contained the non-reference allele. This wouldn't be a total surprise, but I think the article detailed it. But now I can't find the paper again.

Did anyone else see this? Has anyone here tried to quantitate this phenomenon.

thanks in advance.

[jts]

http://bioinformatics.oxfordjournals.org/cgi/content/short/btp579v1

Is this the paper you are looking for?

[nilshomer]

I thought I recently read a paper detailing poorer read mapping to a reference genome if the read contained the non-reference allele. This wouldn't be a total surprise, but I think the article detailed it. But now I can't find the paper again.

Did anyone else see this? Has anyone here tried to quantitate this phenomenon.

thanks in advance.

You can use the "btestindexes" utility in BFAST to test the power, given your alignment settings, of aligning reads with any combination of variants/errors. So to ask what is the difference in power between having 0 SNPs and 1 SNP, find the power using the utility in both cases and compare. If you find that the sensitivity to 1 SNP is too low, then you can easily change the input settings be more sensitive. The same is true for any variant and error combination.

[krobison]

Thanks -- the Bioinformatics paper is definitely the one; it's embarrassing I had to resort to this to find it again.

For the record, the complete cite is

Effect of read-mapping biases on detecting allele-specific expression from RNA-sequencing data.

Jacob F. Degner 1,3,*, John C. Marioni 1,*, Athma A. Pai 1, Joseph K. Pickrell 1, Everlyne Nkadori 1,2, Yoav Gilad 1,* and Jonathan K. Pritchard 1,2,*

1Department of Human Genetics, 2Howard Hughes Medical Institute, and 3Committee on Genetics, Genomics and Systems Biology, University of Chicago, 920 E. 58th St., CLSC 507, Chicago, IL 60637 .

RNA. However, a major technical hurdle lies in the need to map short sequence reads back to their correct locations in a reference genome. Here we investigate the impact of SNP variation on the reliability of read-mapping in the context of detecting allele-specific expression (ASE).

Results: We generated sixteen million 35 bp reads from mRNA of each of two HapMap Yoruba individuals. When we mapped these reads to the human genome we found that, at heterozygous SNPs, there was a significant bias towards higher mapping rates of the allele in the reference sequence, compared to the alternative allele. Masking known SNP positions in the genome sequence eliminated the reference bias but, surprisingly, did not lead to more reliable results overall. We find that even after masking, 5-10% of SNPs still have an inherent bias towards more effective mapping of one allele. Filtering out inherently biased SNPs removes 40% of the top signals of ASE. The remaining SNPs showing ASE are enriched in genes previously known to harbor cis-regulatory variation or known to show uniparental imprinting. Our results have implications for a variety of applications involving detection of alternate alleles from short-read sequence data.

Availability: Scripts, written in Perl and R, for simulating short reads, masking SNP variation in a reference genome, and analyzing the simulation output are available upon request from JFD. Raw short read data were deposited in GEO (http://www.ncbi.nlm.nih.gov/geo/) under accession number GSE18156.