A computational approach to identify point mutations associated with occult hepatitis B: significant mutations affect coding regions but not regulative elements of HBV

[Guest guest]

http://7thspace.com/headlines/391143/a_computational_approach_to_identify_point_\

mutations_associated_with_occult_hepatitis_b_significant_mutations_affect_coding\

_regions_but_not_regulative_elements_of_hbv.html

A computational approach to identify point mutations associated with occult

hepatitis B: significant mutations affect coding regions but not regulative

elements of HBV

Occult Hepatitis B Infection (OBI) is characterized by absence of serum HBsAg

and persistence of HBV-DNA in liver tissue, with low to undetectable serum

HBV-DNA. The mechanisms underlying OBI remain to be clarified.

To evaluate if specific point mutations of HBV genome may be associated with

OBI, we applied an approach based on bioinformatics analysis of complete genome

HBV sequences. In addition, the feasibility of bioinformatics prediction models

to classify HBV infections into OBI and non-OBI by molecular data was evaluated.

Methods: 41 OBI and 162 non-OBI complete genome sequences were retrieved from

GenBank, aligned and subjected to univariable analysis including statistical

evaluation.

Their S coding region was analyzed for Stop codon mutations too, while S amino

acid variability could be evaluated for genotype D only, due to the too small

number of available complete genome OBI sequences from other

genotypes.Prediction models were derived by multivariable analysis using

Logistic Regression, Rule Induction and Random Forest approaches, with

extra-sample error estimation by Multiple ten-fold Cross-Validation (MCV).

Models were compared by t-test on the Area Under the Receiver Operating

Characteristic curve (AUC) distributions obtained from the MCV runs for each

model against the best-performing model.

Results: Variations in seven nucleotide positions were significantly associated

with OBI, and occurred in 11 out of 41 OBI sequences (26.8%): likely, other

mutations did not reach statistical significance due to the small size of OBI

dataset.

All variations affected at least one HBV coding region, but none of them mapped

to regulative elements. All viral proteins, with the only exception of the X,

were affected.

Stop codons in the S, that might account for absence of serum HBsAg, were not

significantly enriched in OBI sequences. In genotype D, amino acid variability

in the S was higher in OBI than non-OBI, particularly in the immunodominant

region.

A Random Forest prediction model showed the best performance, but all models

were not satisfactory in terms of specificity, due to the small sample size of

OBIs; however results are promising in the perspective of a broader dataset of

complete genome OBI sequences.

Conclusions: Data suggest that point mutations rarely occur in regulative

elements of HBV, if ever, and contribute to OBI by affecting different viral

proteins, suggesting heterogeneous mechanisms may be responsible for OBI,

including, at least in genotype D, an escape mutation mechanism due to imperfect

immune control. It appears possible to derive prediction models based on

molecular data when a larger set of complete genome OBI sequences will become

available.

Author: o BruniMattia ProsperiCinzia MarcantonioAlessandra

AmadoriUmbertina VillanoElena TritarelliAlessandra Lo PrestiMassimo Ciccozzi

Ciccaglione

Credits/Source: Virology Journal 2011, 8:394