Comparative Evaluation of Open-Source Bioinformatics Pipelines for Full-Length Viral Genome Assembly

ViralNGSBenchmarking This repository contains supplementary data and code for a study titled "Comparative Evaluation of Bioinformatic Pipelines for Full-Length Viral Genome Assembly," published on BioRxiv (https://doi.org/10.1101/2024.03.13.584779) and in the journal Viruses (https://doi.org/10.3390/v16121824). Data + Configuration/Parametrization To enable easier reproducibility of the analysis, we provide the data and configuration files for all three datasets (SIM, SS+NGS, SGS-FULL) used in our study. Here we preliminarily publish the Sanger and next-generation sequencing results of the 49 Hungarian HIV samples of the SS+NGS dataset. These results will be publicly available in INSDC data repositories (GenBank and SRA) within 1 year from the point of publication in a scientific journal. Until then, the sequencing results uploaded to this repository can only be used for the replication of our benchmarking analysis. Any other use requires the authors' permission. Raw result files Combined raw results of all analyses GABM_NGS_results.csv: Runtime and memory benchmarking results on the NGS (Ode et al.) dataset GABM_SIM_results_revision.csv: Benchmarking results using simulated HIV-1 quasispecies data GABM_SIM_150.csv: Benchmarking results using simulated HIV-1 quasispecies data with 2x150bp paired-end Illumina data GABM_SGS_results.csv: Benchmarking results using the Vibhol et al. SGS dataset GABM_SSNGS_HUN_results.csv: Benchmarking results using Hungarian Sanger and next generation sequencing results from the same samples SGS diversity calculation results Shankarappa.xlsx: Diversity statistics of the Shankarappa et al. dataset Vibholm.xlsx: Diversity statistics of the Vibholm et al. dataset QUAST reports and genome assemblies of the SIM dataset We uploaded all benchmark sequences, genome assemblies and QUAST reports and a subset of simulated sequencing reads (1 replicate/scenario) for our in silico analyses. Scripts The scripts in this repository have been developed to benchmark four bioinformatic pipelines (shiver, SmaltAlign, viralngs, V-Pipe) for full-length viral genome assembly. Main scripts gabm.sh: Main pipeline script organizing different steps of the analysis for one sample. gabm_batch.sh: A wrapper script for gabm.sh. Initializes sample directory for gabm.sh using files in the batch directory. gabm_batch.wrapper.sh: A simple wrapper for gabm_batch.sh to run several different batch analyses after each other. Utilities vngs.sh: Script supplied to the viralngs docker container including all steps of genome assembly with additional runtime and memory benchmarking. align_MAFFT.py: Multiple alignment of .fasta files using MAFFT. changeAmbiguousToA.py: change IUPAC ambiguity codes in Sanger sequences due to software incompatibility before base calling using NGS data. correctSangerByVariants.py: Corrects Sanger sequences in position where NGS data strongly indicates a different base call to provide consistency between datasets. multifastaDiversity.py: Calculation of average pairwise Hamming distances in a multifasta alignment. deleteGapsAlignment.py: Deletes gap characters from multifasta alignments. calculateCoordinatesByMotifs.py: Calculates the coordinates of a genetic interval in a .fasta sequence based on a start and end motif. cropGenomeByCoordinates.py: Crops HIV-1 genomes (.fasta) based on a start and end coordinate on the HXB2 reference genome. changeNBases.py: Changes unidentified bases ('N') to a random nucleotide in a .fasta file. changesToNBases.py: Changes special IUPAC nucleotide ambiguity codes to 'N'. compBamLiftover.py: Compares .bam aligments mapped to highly similar reference sequences. VCF.py: Read in utilities for .vcf file in Python developed by Kamil Slowikowski. VariantAnalysis.py: Compares .vcf variant calls devised from highly similar reference sequences. Scripts for gathering results gather_results.sh: Goes through the folder structure generated by gabm_batch_wrapper.sh and calls gather_results.R in every sample directory. After that, initiates the merging of results by merge_results.R gather_results.R: Collects the benchmarking results in every sample directory. merge_results.R: Merges the tables generated by gather_results.R. Extra scripts SGS_rename.sh: Convenience function for renaming .fasta files. ss_ngs_download.sh: Convenience function for downloading data from NCBI Nucleotide and the Sequence Read Archive.ViralNGSBenchmarking 本仓库为题为《全长病毒基因组组装生物信息学流程对比评估》的研究提供补充数据与代码，该研究已发表于生物预印本服务器（BioRxiv，https://doi.org/10.1101/2024.03.13.584779）及期刊《Viruses》（https://doi.org/10.3390/v16121824）。 数据与配置/参数化设置 为便于研究结果复现，我们提供了本研究中使用的全部三个数据集（SIM、SS+NGS、SGS-FULL）的数据与配置文件。此处我们先期发布SS+NGS数据集内49份匈牙利HIV样本的桑格测序与下一代测序结果。上述测序结果将在期刊发表后1年内，于国际核酸序列数据库联盟（INSDC）的仓储库（基因银行GenBank与序列读取档案SRA）中公开。在此之前，本仓库上传的测序结果仅可用于复现本基准测试分析，其他用途需征得作者许可。 原始结果文件 所有分析的合并原始结果 GABM_NGS_results.csv：NGS（Ode等）数据集上的运行时与内存基准测试结果 GABM_SIM_results_revision.csv：基于模拟HIV-1准种数据的基准测试结果 GABM_SIM_150.csv：基于2×150bp双端Illumina测序数据的模拟HIV-1准种数据的基准测试结果 GABM_SGS_results.csv：基于Vibhol等的SGS数据集的基准测试结果 GABM_SSNGS_HUN_results.csv：基于同一批样本的匈牙利桑格测序与下一代测序结果的基准测试结果 SGS多样性计算结果 Shankarappa.xlsx：Shankarappa等数据集的多样性统计数据 Vibholm.xlsx：Vibholm等数据集的多样性统计数据 SIM数据集的QUAST报告与基因组组装结果 我们上传了本次虚拟分析中使用的全部基准序列、基因组组装结果、QUAST报告，以及部分模拟测序读长（每个场景1个重复）。 脚本 本仓库开发的脚本用于对四款生物信息学流程（shiver、SmaltAlign、viralngs、V-Pipe）进行全长病毒基因组组装的基准测试。 主脚本 gabm.sh：用于单样本分析各步骤组织的主流程脚本。 gabm_batch.sh：gabm.sh的包装脚本，利用批次目录内的文件为gabm.sh初始化样本目录。 gabm_batch.wrapper.sh：gabm_batch.sh的简易包装器，可依次运行多个不同批次的分析任务。 实用工具脚本 vngs.sh：用于viralngs容器的脚本，包含基因组组装的全部步骤，并额外集成运行时与内存基准测试功能。 align_MAFFT.py：利用MAFFT对.fasta格式文件进行多序列比对。 changeAmbiguousToA.py：因软件兼容性问题，将桑格测序序列中的IUPAC歧义碱基代码替换为腺嘌呤（A），以适配后续NGS数据的碱基识别流程。 correctSangerByVariants.py：基于NGS数据明确指示的碱基差异，校正桑格测序序列的对应位点，以保证不同数据集间的一致性。 multifastaDiversity.py：计算多fasta比对文件中的平均成对汉明距离。 deleteGapsAlignment.py：从多fasta比对文件中移除间隙字符。 calculateCoordinatesByMotifs.py：基于起始与终止基序，在.fasta序列中计算指定遗传区间的坐标。 cropGenomeByCoordinates.py：基于HXB2参考基因组上的起始与终止坐标，裁剪HIV-1基因组的.fasta文件。 changeNBases.py：将.fasta文件中的未识别碱基（'N'）替换为随机核苷酸。 changesToNBases.py：将特殊IUPAC核苷酸歧义代码转换为'N'。 compBamLiftover.py：比对映射至高度相似参考序列的.bam格式文件。 VCF.py：由Kamil Slowikowski开发的Python版.vcf格式文件读取工具。 VariantAnalysis.py：比对基于高度相似参考序列得到的.vcf格式文件变异调用结果。 结果收集脚本 gather_results.sh：遍历gabm_batch_wrapper.sh生成的文件夹结构，调用每个样本目录内的gather_results.R，随后通过merge_results.R启动结果合并流程。 gather_results.R：收集每个样本目录内的基准测试结果。 merge_results.R：合并gather_results.R生成的结果表格。 额外脚本 SGS_rename.sh：用于重命名.fasta文件的便捷工具脚本。 ss_ngs_download.sh：用于从NCBI核苷酸数据库与序列读取档案下载数据的便捷工具脚本。