Reliable variant calling during runtime of Illumina sequencing

Abstract

The sequential paradigm of data acquisition and analysis in next-generation sequencing leads to high turnaround times for the generation of interpretable results. We combined a novel real-time read mapping algorithm with fast variant calling to obtain reliable variant calls still during the sequencing process. Thereby, our new algorithm allows for accurate read mapping results for intermediate cycles and supports large reference genomes such as the complete human reference. This enables the combination of real-time read mapping results with complex follow-up analysis. In this study, we showed the accuracy and scalability of our approach by applying real-time read mapping and variant calling to seven publicly available human whole exome sequencing datasets. Thereby, up to 89% of all detected SNPs were already identified after 40 sequencing cycles while showing similar precision as at the end of sequencing. Final results showed similar accuracy to those of conventional post-hoc analysis methods. When compared to standard routines, our live approach enables considerably faster interventions in clinical applications and infectious disease outbreaks. Besides variant calling, our approach can be adapted for a plethora of other mapping-based analyses.

Deep learning-based real-time detection of novel pathogens during sequencing

“…To estimate the sample size that can be analyzed in real-time after parsing or mapping with HiLive2, we measured how many reads per second could be processed by the pathogenicity prediction methods compared in this study. Then, we calculated the number of predictions feasible in a time-frame corresponding to 25 cycles (with wall-time per cycle as in (Loka et al, 2019)). Note that this is an inherently difficult comparison, as inference with deep learning models can be accelerated with GPUs, while other methods cannot.…”

“…However, lower throughput and high error rates of those technologies impede their adoption for pathogen detection. Scalability, cost-efficiency and accuracy of Illumina sequencing still make it a gold standard, although it may change in the future with the establishment of improved ONT protocols and computational methods (Loka et al ., 2019).…”

“…An alternative approach is to use the general purpose computational infrastructure and optimize the algorithms for fast and accurate analysis of incomplete reads as they are produced, during the sequencing run. HiLive (Lindner et al ., 2017) and HiLive2 (Loka et al ., 2019) are real-time mappers, performing on par with the traditional mappers like Bowtie2 (Langmead and Salzberg, 2012) and BWA (Li and Durbin, 2010) with no live-analysis capabilities. However, as read mappers are designed for fast and precise sequence alignment, they are expected to miss most of the reads originating from genomes highly divergent from the available references.…”

ApharSeq: An Extraction-free Early-Pooling Protocol for Massively Multiplexed SARS-CoV-2 Detection

“…While a 12-18 hour turnaround time is probably prohibitive for emergency testing, we believe that for large scale and routine population screens it is reasonable. Further, there are potential optimizations and workarounds that can reduce the sequencing runtime by a factor of 2-3 22 .…”

DeePaC: Predicting pathogenic potential of novel DNA with a universal framework for reverse-complement neural networks

“…Read mappers, for example BWA (Li and Durbin, 2009) and Bowtie2 (Langmead and Salzberg, 2012) fall into this category. Live mapping approaches such as HiLive and HiLive2 (Lindner et al, 2017;Loka et al, 2018) can even map the reads in real time, as the sequencer is running, leading to a drastic reduction in total analysis time.…”