ExonDepth github
ExonDepth 帮助文档
ExomeDepth 是一个 R 软件包，旨在使用高通量 DNA 序列数据检测遗传拷贝数变异 (CNV)。虽然外显子组包含在包的名称中，但实际上它在较小的面板上表现最佳，因为包的分析利用了（通常）并行运行的大量样本之间的紧密相关结构。这些紧密的相关性正是 ExomeDepth 在为每个测试样本构建参考样本时所寻找的（默认要求对照集中需要存在一个相关性大于0.97的样本，否则失败），并且输出的质量通常会根据相关性结构而变化。
| Tool | Version | Language | Availability | Methods | Number of evaluated parameters | Year (paper publication) | Citationsa | PMID | Bench- marked in [14] |
|---|---|---|---|---|---|---|---|---|---|
| Atlas-CNV | 0 | R and Perl program | https://github.com/theodorc/Atlas-CNV | It normalizes individual read depth data to average read depth per target, converting it to reads per kilobase million (RPKM). It computes log2 scores for each sample/median ratio at every exon, assessing sample quality via SampleQC, checking StDev of log2 scores and analysis of variance (ANOVA) on mean RPKM coverage. | 2 | 2019 | 14 | 30890783 | No |
| ClearCNV | 0.306 | Python program | https://github.com/bihealth/clear-cnv | It utilizes match scores to group samples based on coverage patterns. It employs data normalization, scaled z-scores, and r-scores to identify copy number variations (CNVs) in both multi-exon and single-exon regions. | 7 | 2022 | 1 | 35751599 | No |
| ClinCNV | 1.18.3 | R, Java, Python program | https://github.com/imgag/ClinCNV | ClinCNV employs an algorithm that combines the strengths of circular binary segmentation and hidden Markov model–based techniques to perform multi-sample normalization and CNV calling. | 2 | 2022b | 6 | – | No |
| CNVkit | 0.9.10 | Python program | https://github.com/etal/cnvkit | It uses targeted and the nonspecifically captured off-target reads to calculate log2 copy ratios across the genome. | 18 | 2016 | 1212 | 27100738 | No |
| Cobalt | 0.8.0 | Python program | https://github.com/ARUP-NGS/cobalt | It introduces two algorithmic adaptations to improve accuracy in a hidden Markov model. A method for computing target and copy number–specific emission distributions and they perform pointwise maximum posteriori HMM decoding to improve sensitivity for small CNV. | 8 | 2022 | 0 | 35854218 | No |
| CODEX2 | 1.3.00 | R package | https://github.com/yuchaojiang/CODEX2 | Based on CODEX package, it models the GC content bias and normalizes the read depth data for CNV detection via a Poisson latent factor model. | 8 | 2018 | 39 | 30477554 | Yes (v.1.2.0) |
| CoNVaDING | 1.2.1 | Perl program | https://github.com/molgenis/CoNVaDING | Combination of ratio scores and Z-scores of the sample of interest compared to the selected normalized control samples. | 7 | 2016 | 67 | 26864275 | Yes (v.1.2.0) |
| DECoN | 2.0.1 | R program | https://github.com/RahmanTeam/DECoN | Modifies ExomeDepth package by altering the hidden Markov model probabilities to depend upon the distance between exons. | 3 | 2016 | 59 | 28459104 | Yes (v.1.0.1) |
| ExomeDepth | 1.1.16 | R package | https://github.com/vplagnol/ExomeDepth | Beta-binomial model with GC correction and hidden Markov model to combine likelihood across exons. | 3 | 2012 | 516 | 22942019 | Yes (v.1.1.10) |
| GATK-gCNV | 4.5.00 | Java, Python, R program | https://github.com/broadinstitute/gatk | It calculates read counts over specified genomic regions per sample; it clusters technically similar samples using principal component analysis to reduce biases and enhance efficiency. After estimating chromosomal ploidy, it denoises read depth, infers CNVs via a unified model using the Viterbi algorithm | 35 | 2023 | 0 | 37604963 | No |
| pan-elcn.MOPS | 1.20.00 | R package | https://github.com/bioinf-jku/panelcn.mops | Adaptation of cn.MOPS package, which decomposes variations in coverage across samples into integer copynumbers and noise by means of its mixture components and Poisson distributions. | 13 | 2017 | 53 | 28449315 | Yes (v.1.0.0) |
| VisCap | 0.8 | R program | https://github.com/pughlab/VisCap | It determines the portion of sequence coverage allocated to genomic intervals and calculates log2 ratios compared to the median of reference samples with a matching test setup.CNV candidates are identified when log2 ratios surpass thresholds set by the user. | 2 | 2016 | 49 | 26681316 | No |