Semi-artificial datasets as a resource for validation of bioinformatics pipelines for plant virus detection

• DOI: 10.24072/pcjournal.62
• Publication Date: 2021-12-02T07:48:48Z
• Authors: Lucie Tamisier; Annelies Haegeman; Yoika Foucart; Nicolas Fouillien; Maher Al Rwahnih; Nihal Buzkan; Thierry Candresse; Michela Chiumenti; Kris De Jonghe; Marie Lefebvre; Paolo Margaria; Jean Sébastien Re...; Kristian Stevens; Denis Kutnjak; Sébastien Massart
• Keywords: 0301 basic medicine; 570; 0303 health sciences; Bioinformatics; [SDV]Life Sciences [q-bio]; Science; Plant virus detection; Bioinformatics pipelines; Q; Semi-artificial dataset; Benchmark; 630; [SDV.BV.PEP]Life Sciences [q-bio]/Vegetal Biology/Phytopathology and phytopharmacy; 3. Good health; High-Throughput Sequencing; 03 medical and health sciences; Reference data; Archaeology; [SDV.BV]Life Sciences [q-bio]/Vegetal Biology; NA; HTS; Haplotype reconstruction; CC1-960

The widespread use of High-Throughput Sequencing (HTS) for detection plant viruses and sequencing virus genomes has led to the generation large amounts data bioinformatics challenges process them. Many pipelines are available, making choice a suitable one difficult. A robust benchmarking is needed unbiased comparison pipelines, but there currently lack reference datasets that could be used this purpose. We present 7 semi-artificial composed real RNA-seq from virus-infected plants spiked with artificial reads. Each dataset addresses prevent detection. also 3 showing challenging composition as well 8 completely test haplotype reconstruction software. With these address several diagnostic challenges, we hope encourage virologists, diagnosticians bioinformaticians evaluate benchmark their pipeline(s).