Mancuso Nicholas, Tork Bassam, Skums Pavel, Ganova-Raeva Lilia, Măndoiu Ion, Zelikovsky Alex
Department of Computer Science, Georgia State University, Atlanta, GA, USA.
In Silico Biol. 2011;11(5-6):237-49. doi: 10.3233/ISB-2012-0458.
This paper addresses the problem of reconstructing viral quasispecies from next-generation sequencing reads obtained from amplicons (i.e., reads generated from predefined amplified overlapping regions). We compare the parsimonious and likelihood models for this problem and propose several novel assembling algorithms. The proposed methods have been validated on simulated error-free HCV and real HBV amplicon reads. The new algorithms have been shown to outperform the method of Prosperi et. al. Our experiments also show that viral quasispecies can be reconstructed in most cases more accurately from amplicon reads rather than shotgun reads. All algorithms have been implemented and made available at https://bitbucket.org/nmancuso/bioa/.
本文探讨了从扩增子（即从预定义的扩增重叠区域生成的 reads）获得的下一代测序 reads 中重建病毒准种的问题。我们比较了针对此问题的简约模型和似然模型，并提出了几种新颖的组装算法。所提出的方法已在模拟的无误差 HCV 和真实 HBV 扩增子 reads 上得到验证。新算法已被证明优于 Prosperi 等人的方法。我们的实验还表明，在大多数情况下，从扩增子 reads 而不是鸟枪法 reads 中能更准确地重建病毒准种。所有算法均已实现，并可在 https://bitbucket.org/nmancuso/bioa/ 上获取。
