|
Accuracy testing of deep sequencing for HIV-1 drug resistance testing in the QCMD 2015 ENVA HIV drug resistance typing |
|---|---|
| รหัสดีโอไอ | |
| Creator | Pornpimon Nimitsuntiwong, Chorthip Wathiphaba, Ekawat Pasomsab, Wasun Chantratita |
| Title | Accuracy testing of deep sequencing for HIV-1 drug resistance testing in the QCMD 2015 ENVA HIV drug resistance typing |
| Publisher | Genetics Society of Thailand |
| Publication Year | 2559 |
| Journal Title | Genomics and Genetics |
| Journal Vol. | 9 |
| Journal No. | 2&3 |
| Page no. | 104-109 |
| Keyword | HIV drug resistance, NGS technologies, deep sequencing, QCMD ENVA HIV drug resistance |
| ISSN | 24655198 |
| Abstract | Sanger sequencing which is a gold standard method for genotypic drug resistance testing has limited sensitivity in detecting HIV-1 drug resistance mutations (DRMs) at frequencies below 20% of viral quasispecies. Deep sequencing is an ultrasensitive method, which allows detecting such mutations and mutations detected by Sanger sequencing. A newly developed HIV-1 deep sequencing drug resistance assay has never been tested the accuracy with external quality assessment (EQA) program. The objective of this study was to test the accuracy of deep sequencing in detecting DRMs in HIV-1 protease (PR) and reverse transcriptase (RT) genes with five samples from ENVA 2015 HIV drug resistance typing EQA program. According to the 2015 EQA program report, deep sequencing could generate complete datasets of all five ENVA15 panel samples, which covered both PR and RT genes and comprised all IAS codons. In addition, deep sequencing detected 339 of 340 DRM codons and was awarded 99.86% of the overall sequence concordance. A manual reviewed bam file of ENVA15-08 sample was performed to investigate the incorrect codon and found at position 2,381 (PR-43) had a mixture of A (wild-type) at 83% and G (mutant type) at 16% which was identical to the expected results (R). Therefore, we reanalyzed the sequences of all samples by another pipeline found deep sequencing detected 334 DRM codons identical to expected results which comprised the incorrect codon and 333 DRM codons. In addition, the other 6 codons which comprised mutants at frequencies below 20% of viral quasispecies were partial concordance with the expected results because 2015 ENVA consensus sequences were created by aligning sequences summited by all participants whose almost all datasets were based on Sanger sequencing technology. In conclusion, deep sequencing has accuracy in detecting HIV-1 DRMs and would be adopted as a clinical laboratory routine in the near future. |
