Show simple item record

dc.contributor.advisorMullins, James Ien_US
dc.contributor.authorIyer, Shyamalaen_US
dc.date.accessioned2015-09-29T21:25:48Z
dc.date.submitted2015en_US
dc.identifier.otherIyer_washington_0250E_14403.pdfen_US
dc.identifier.urihttp://hdl.handle.net/1773/34059
dc.descriptionThesis (Ph.D.)--University of Washington, 2015en_US
dc.description.abstractMassively parallel sequencing technologies have been extensively applied in HIV-1 research to study the presence of minority variants. Insight gained through these technologies includes identification of minor drug resistant variants and immune escape variants. However, given the massive amounts of data generated, processing the sequences and discerning true minor variants from sequencing artifacts is important. Additionally, errors introduced during viral template amplification and incorrect quantification of templates prior to the sequencing process can further obfuscate resolving mismatch errors within the sequences. I address these issues in this dissertation. We developed a computational algorithm, CorQ, to correct specific patterns of sequencing errors and call Single Nucleotide Polymorphisms (SNPs). When coupled with additional error correction steps, we observed a 97% reduction in insertion, and deletion sequencing errors. In addition, we observed over 98% specificity in SNP detection compared to other available error correction methods. We observed reduced SNP calling specificity when error correction programs were tested on sequences with simulated PCR amplification mismatch errors, with the highest specificity of 70% observed with a combination CorQ algorithm, highlighting the difficulty in resolving errors generated during PCR amplification. We observed over 99% concordance in consensus variants observed in multiple HIV-1 infected subjects sequenced with traditional Sanger sequencing and pyrosequencing. The majority of SNPs that were specific to subjects’ pyrosequences were present at less than 2% of the subjects viral sequence population. We observed higher accuracy in variant frequencies in positions where read coverage exceeded the number of input templates. We have applied the developed error correction algorithm and observations from SNP variant comparisons to identifying major and minor variants observed within predicted T-cell epitope regions in HIV-1 infected study subjects enrolled in the MRKAd5 STEP vaccine trial. We observed genetic signatures of immune responses primed by the vaccine on breakthrough HIV-1 sequences. We observed greater genetic distances to the vaccine sequence in breakthrough sequences from vaccine recipients than placebo recipients and this difference was most significant within T-cell epitope regions. Additionally over time, the vaccine-primed immune responses resulted in reduced epitope diversity and decreased rates of epitope evolution over time. Combined, these results strongly support the hypothesis that the MRKAd5 vaccine resulted in T-cell mediated selection occurring post-infection. The results from our study are the first evidence of vaccine-induced anamnestic pressure influencing CTL epitope evolution and epitope diversity over time during HIV-1 infection.en_US
dc.format.mimetypeapplication/pdfen_US
dc.language.isoen_USen_US
dc.rightsCopyright is held by the individual authors.en_US
dc.subjectGenetic signatures; HIV-1 vaccine; immune responses; Pyrosequencesen_US
dc.subject.otherVirologyen_US
dc.subject.otherBioinformaticsen_US
dc.subject.othermicrobiologyen_US
dc.titleIdentifying genetic signatures of vaccine-induced immune responses in HIV-1 infected MRKAd5 STEP vaccine study subjectsen_US
dc.typeThesisen_US
dc.embargo.termsRestrict to UW for 2 years -- then make Open Accessen_US
dc.embargo.lift2017-09-18T21:25:48Z


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record