Wednesday, October 10, 2012

Fwd: SpolPred: Rapid and accurate prediction of Mycobacterium tuberculosis spoligotypes from short genomic sequences

Fwd: please follow footer link
SpolPred: Rapid and accurate prediction of Mycobacterium tuberculosis spoligotypes from short genomic sequences:

Summary: Spoligotyping is a well-established genotyping technique based on the presence of unique DNA sequences in Mycobacterium tuberculosis (Mtb), the causal agent of tuberculosis disease (TB). Whilst advances in sequencing technologies are leading to whole genome bacterial characterisation, tens of thousands of isolates have been spoligotyped, giving a global view of Mtb strain diversity. To bridge the gap, we have developed SpolPred, software to predict the spoligotype from raw sequence reads. Our approach is compared with experimentally and de novo assembly determined strain-types in a set of 44 Mtb isolates. In silico and experimental results are identical for almost all isolates (39/44). However, SpolPred detected 5 experimentally false spoligotypes, and was more accurate and faster than the assembling strategy. Application of SpolPred to an additional 7 isolates with no laboratory data, led to types that clustered with identical experimental types in a phylogenetic analysis using SNPs. Our results demonstrate the usefulness of the tool and its role in revealing experimental limitations.

Availability and implementation:SpolPred is written in C and is available from www.pathogenseq.org/spolpred.

Contact:francesc.coll@lshtm.ac.uk



(Original Post: Bioinformatics - Advance Access.)