Simulated NGS read datasets for novel human virus prediction

This repository contains simulated Illumina read datasets for novel human virus prediction and associated metadata extracted from the Virus Host Database (https://www.genome.jp/virushostdb/). The reads are 250bp long and were simulated with Mason (https://www.seqan.de/apps/mason/) from genomes downloaded from NCBI. The training-validation-test split was done on whole viral sequences to ensure "novelty" of validation and test viruses. The training sets contain 10 million reads per class, validation sets - 1.25 million reads per class, and test sets - 1.25 million paired reads per class. The negative class sets contain reads simulated from chordate-infecting ("cho"), metazoan-infecting ("met"), eukariote-infecting ("euk") and all-nonhuman viruses. The positive class contains human-infecting viruses. The stratified dataset ("strat") contains an equal number of reads from "cho", "met but not cho", "euk but not met" and "all but not euk". 

Species-level datasets ("humspec", "allspec" and "chospec", with the corresponding fasta and *_species.rds files) are constructed analogously, but ensuring that all viruses of a given species were assigned to either training, val or test set. This is a stricter setting modelling a "novel viral species" scenario while reflecting within-species phenotype diversity.

blast_hits.gz contains blast hits of human virome reads form Moustafa et al., 2017 (https://doi.org/10.1371/journal.ppat.1006292) blasted against our training database (see paper for details). In the second column you can find the matched label and the accession number of the matched reference. blast_labels_complete.gz contains extracted labels for all virome reads, including those without any matches. Note: one of the read headers (>3c8ac47039d32b11c8fe23f588e444e9) from Moustafa et al. is slightly corrupted with null characters. You can remove them with sed 's/\x0//g' or equivalent.

本仓库收录了用于新型人类病毒预测的模拟Illumina测序读段数据集，以及从病毒宿主数据库（Virus Host Database，https://www.genome.jp/virushostdb/）中提取的关联元数据。该测序读段长度为250bp，通过Mason测序模拟器（https://www.seqan.de/apps/mason/），基于从NCBI下载的病毒基因组模拟生成。训练-验证-测试划分以完整病毒序列为单位进行，以确保验证集与测试集的病毒具备“新颖性”。训练集每类包含1000万条读段，验证集与测试集每类分别包含125万条读段与125万条配对读段。负样本类别包含从感染脊索动物（"cho"）、感染后生动物（"met"）、感染真核生物（"euk"）以及所有非人类病毒中模拟得到的读段；正样本类别则包含感染人类的病毒。分层数据集（"strat"）中，"感染脊索动物（cho）"、"感染后生动物但非脊索动物（met but not cho）"、"感染真核生物但非后生动物（euk but not met）"以及"所有非真核生物感染病毒（all but not euk）"的读段数量保持均等。

物种级数据集（"humspec"、"allspec"与"chospec"，对应配套文件为fasta格式及*_species.rds文件）的构建逻辑与上述类似，但要求同一物种的所有病毒均被划分至训练、验证或测试集中的单一集合。该设置更为严格，可模拟“新型病毒物种”场景，同时兼顾物种内的表型多样性。

blast_hits.gz文件包含来自Moustafa等人2017年研究（https://doi.org/10.1371/journal.ppat.1006292）的人类病毒组读段，与本仓库训练数据库进行BLAST比对后的结果（详细信息参见论文），其第二列包含匹配到的标签以及对应参考序列的登录号。blast_labels_complete.gz文件则包含了所有病毒组读段的提取标签，其中也包含无任何匹配结果的读段。注意：Moustafa等人研究中的一条读段头部（>3c8ac47039d32b11c8fe23f588e444e9）存在空字符损坏，可通过`sed 's/x0//g'`或等效命令修复。