Proficiency Testing of Virus Diagnostics Based on Bioinformatics Analysis of Simulated In Silico High-Throughput Sequencing Data Sets

Abstract

Quality management and independent assessment of high-throughput sequencing-based virus diagnostics have not yet been established as a mandatory approach for ensuring comparable results. The sensitivity and specificity of viral high-throughput sequence data analysis are highly affected by bioinformatics processing using publicly available and custom tools and databases and thus differ widely between individuals and institutions. Here we present the results of the COMPARE [Collaborative Management Platform for Detection and Analyses of (Re-)emerging and Foodborne Outbreaks in Europe] in silico virus proficiency test. An artificial, simulated in silico data set of Illumina HiSeq sequences was provided to 13 different European institutes for bioinformatics analysis to identify viral pathogens in high-throughput sequence data. Comparison of the participants’ analyses shows that the use of different tools, programs, and databases for bioinformatics analyses can impact the correct identification of viral sequences from a simple data set. The identification of slightly mutated and highly divergent virus genomes has been shown to be most challenging. Furthermore, the interpretation of the results, together with a fictitious case report, by the participants showed that in addition to the bioinformatics analysis, the virological evaluation of the results can be important in clinical settings. External quality assessment and proficiency testing should become an important part of validating high-throughput sequencing-based virus diagnostics and could improve the harmonization, comparability, and reproducibility of results. There is a need for the establishment of international proficiency testing, like that established for conventional laboratory tests such as PCR, for bioinformatics pipelines and the interpretation of such results.

Perspective on Proteomics for Virus Detection in Clinical Samples

“… 14 The advantage of NGS is that no prior knowledge of the target virus is necessary, but detection limits may vary dramatically. 15 Another method called AmpliSeq combines PCR with NGS, allowing targeted virus detection with higher sensitivity and multiplexing capacity. 16 Protein-based methods most often rely on antibodies for virus detection.…”

Next-generation diagnostics: virus capture facilitates a sensitive viral diagnosis for epizootic and zoonotic pathogens including SARS-CoV-2

“…The combination of a generic and manipulation-free HTS workflow with the capture approach leads to an enhancement of the viral signal in terms of sequence reads in the sequence dataset resulting in higher sensitivity of HTS. The identification of distantly related viruses can be difficult as shown for a deviating avian bornavirus in a proficiency test for HTS bioinformatics [29]. Although the problem in the mentioned study was related to bioinformatics analyses, the accumulation of certain virus reads by capture enrichment can avoid overlooking pathogenic viruses contained in a sample.…”

PACIFIC: A lightweight deep-learning classifier of SARS-CoV-2 and co-infecting RNA viruses

“…However, the use of RNA-seq in diagnostic settings is often complicated due to complex analytical workflows (34,42). A typical workflow for virus detection in high-throughput sequencing data involves quality assessment and filtering of raw data, removal of host sequences, de novo assembly of remaining reads, and lastly, the alignment and annotation of the generated contigs (43). Implementation of these workflows require expert knowledge of bioinformatics software and databases and often dedicated computing facilities.…”