PePApipe: complete bioinformatics analysis pipeline for African Swine Fever Virus (ASFV) genome v1

African Swine Fever Virus (ASFV) is of high concern in porcine livestock across the world due to both the high mortality associated with this disease in pigs and the trade barriers put in place on regions or countries affected by it. The viral genome is large and complex, but its analysis is of outmost relevance to trace the origin and evolution of the virus. Although several tools exist than can be used to perform assembly and analysis of the DNA genome of ASFV, the authors of this work have not found to date a tool able to compile all steps involved in this analysis in a simple and systematically ordered way. Hence, they have developed a custom-built tool called ´PePApipe´ that is fit for purpose in implementing this work in a rapid, complete, efficient and clear manner. This pipeline has been developed bearing in mind that end-users are mostly laboratory professionals with limited bioinformatics skills, although some basic bioinformatics commands are necessary. Starting from raw data (paired .FASTQ files) obtained from short-read sequencing platforms (Illumina), ‘PePApipe’ implements in a sequential manner all crucial steps by using 13 software tools adequately built into a single pipeline. This pipeline, designed and programmed in Python, can be run locally or via a bash script on a slurm protocol and can be executed on multi-sample batches. The work is performed over three main areas of genomic analysis: quality control and pre-processing of raw reads, de novo genome assembly and variant calling. The algorithm has been designed to be executed in one go, assuming that the amount (%) of ASFV reads in the original .FASTQ files is enough to allow for this (typically occurring from sequencing strategies incorporating viral enrichment steps). However, if the amount (%) of ASFV reads is below (e.g. <75%), the pipeline includes a parallel execution lane designed to produce a new set of .FASTQ files containing only reads belonging to viruses. The two main outputs of the pipeline are the DNA consensus sequence of the ASFV genome and a file with a list of its putative variants (SNPs, insertions and deletions), from the chosen reference genome sequence/s. The pipeline produces all files necessary to interpret the results, classifying them into folders, in an approximate time which depends on the size of the .FASTQ files. The time used may span from 10 minutes with files of around 0.5 Mb through to 30 minutes with files of 1 Mb in size or to longer times if high capacity sequencers are used. Several intermediate files are also created in the process, which can be used as inputs for further or parallel analyses, as well as to check for possible errors during the analysis. All steps are amenable to user control by means of switching on/off the necessary sections in each particular case. There are several steps where the quality of the processes can be checked to make sure the progress of the results is in line with our expectations. In addition, the pipeline can be easily adapted to viruses other than ASFV by changing the parameters relevant to the new virus and running the specific pipeline sections accordingly. This pipeline is adaptable and complementary to post-analysis steps such as viral genome annotation or genome visualization. A complete, reliable and standard assembly sequence is essential in viral genomics as a starting point prior to building on further steps of analysis such as phylogenetic studies, genome-wide association or machine learning/ deep learning/ artificial intelligence studies. An additional advantage of executing a single structured pipeline for viral genome analysis as a block (or in two blocks when the amount of viral reads is lower than 75%, as described above) is that the introduction of errors by the user is minimal, which is not the case when inputs and outputs are handled multiple times, thus increasing the likelihood and/or accumulation of errors. The future plan is to include a viral genome annotation tool in the pipeline to harmonize this crucial and complex step for a large virus which can encompass more than 180 ORFs along the genome.