Srfim2: Using Basecalling Model Parameter Estimates to Understand Sequencing Bias

• Activity Number: 631
• Type: Invited
• Date/Time: Thursday, August 2, 2012 : 10:30 AM to 12:20 PM
• Sponsor: ENAR
• Author(s): Hector Corrada Bravo*+ and Joyce Hsiao
• Companies: University of Maryland and University of Maryland
• Keywords: Genomics ; Second-generation sequencing ; Base-calling

Abstract

Second-generation sequencing technology has revolutionized high-throughput genomics in life science and clinical research. The scale of sequence generated by these instruments has allowed unprecedented views into a number of molecular phenomena, incuding population genetics, transcriptomics, epigenetics, and translational profiling. In the last three years, both the throughput and accuracy of secgen instruments has increased significantly. Nonetheless, there is an increased awareness of the fact that error-rates are not negligible in these technologies, and more importantly, they depend in complex ways on technical and biochemical features of the instrument and sequencing library. Using metrics derived from parameter estimates given by Srfim, our statistical method for base-calling and quality assessment, we quantify biases resulting from sequence-context and nucleotide composition.