Alignment of DNA mass-spectral profiles using network flows

Abstract

Mass spectrometry (MS) of DNA fragments generated by base-specific cleavage of PCR products emerges as a cost-effective and robust alternative to DNA sequencing. MS has been successfully applied to SNP discovery using reference sequences, genotyping and detection of viral transmissions. Although MS is yet to be adapted for reconstruction of genetic composition of complex intra-host viral populations on the scale comparable to the next-generation DNA sequencing technologies, the MS profiles are rich sources of data reflecting the structure of viral populations and completely suitable for accurate assessment of genetic relatedness among viral strains. However, owing to a data structure, which is significantly different from sequences, application of MS profiles to genetic analyses remains a challenging task. Here, we develop a novel approach to aligning DNA MS profiles and assessment of genetic relatedness among DNA species using spectral alignments (MSA). MSA was formulated and solved as a network flow problem. It enables an accurate comparison of MS profiles and provides a direct evaluation of genetic distances between DNA molecules without invoking sequences. MSA may serve as accurately as sequence alignments to facilitate phylogenetic analysis and, as such, has numerous applications in basic research, clinical and public health settings. © 2013 Springer-Verlag.