Download The novel virus genotyping tool (STAR) and its use in subtyping

The novel virus genotyping tool (STAR) and its use in
subtyping Human Immunodeficiency Virus Type-1 and
Hepatitis B viruses.
Richard Myers1, Catherine V. Gale1, Caroline Clark1, Julie Bennett1, Richard S.
Tedder1, Ian G. Williams2 and Paul Kellam1
Department of Infection1 and Department of Sexually Transmitted Diseases2, University College
London.
We have developed a high throughput computational tool for assigning a genotype
(subtype) to medically important human viruses Human Immunodeficiency virus type
1 (HIV-1) and Hepatitis B virus (HBV). The HIV subtyping tool works on protease
and reverse transcriptase (PR-RT) amino acid sequence, generated routinely for
detecting genotypic drug resistance whereas the HBV tool works on the whole virus
genome. Subtype specific profiles were created by generation of position specific
scoring matrices (PSSMs) from multiple amino acids or nucleotide alignments of
HIV-1 and HBV sequence data from GenBank, divided into subtypes A, AG, B, C, D,
F, G, H, J and K, and the separate groups N and O for HIV and subtypes A-H for
HBV. Sequences of unknown subtype are aligned with these profiles and a discrete
odds ratio score is derived from variant amino acid (or nucleotide) positions in the
unknown sequence, compared to the normalized frequency distribution of amino acids
(or nucleotides) at the corresponding position in the subtype alignment. The highest
score is used to assign subtype to the query sequence. Leave one out cross validation
analysis showed Subtype Analyser (STAR) was 98%(HIV-1) to 100%(HBV) accurate
in subtype assignation and performed as well, or better, compared to other publicly
available subtyping methods. STAR was used to classify HIV-1 PR-RT sequences
from 843 HIV-1 clinical isolates submitted for drug resistance profiling in London,
885 HIV-1 sequences from Birmingham and from 71 sequences obtained from the
Caribbean. Within the London and Birmingham cohorts 28% and 21% of sequences
respectively were classified by STAR as non-B subtypes, whereas within the
Caribbean dataset, 10% were classified as non-B. The accuracy and validated scoring
mechanism of STAR will allow the investigation of the impact of virus subtypes on
clinical outcomes of infection.