Data and code for "Predicting the functional impact of single nucleotide variants in Drosophila melanogaster with FlyCADD"

Purpose Single nucleotide polymorphisms (SNPs), the most common form of genomic variation, play key roles in micro-evolution and adaptation. In Drosophila melanogaster, many SNPs have been associated to phenotypes through association studies, yet functional validation remains challenging and experimental evidence for functional impact rare. Here, we present FlyCADD, an impact prediction tool that integrates high-quality D. melanogaster genome annotations into a single score reflecting the predicted impact of a SNP. FlyCADD can be applied to distinguish causal from neutral variants, 1) for variant ranking and prioritization of SNPs for functional studies, 2) to improve genome-editing experimental design or evaluation and 3) to enhance interpretation of naturally occuring SNPs, thereby improving our understanding of genotype-phenotype relationships. Dataset content We provide FlyCADD impact prediction scores readily available, both as precomputed scores for all possible single nucleotide variants on the D. melanogaster reference genome and through a locally executable pipeline for scoring novel variants of interest. If you want the FlyCADD scores to your SNPs of interest, they can often be found in the precomputed FlyCADD score files without the need to run the pipeline. If you have any questions, feel free to contact [e-mail]. This repository provides: Precomputed FlyCADD scores for all possible SNPs on the D. melanogaster reference genome Release 6 (.csv files) A locally executable FlyCADD pipeline for scoring novel variants, including the trained logistic regression model files, annotation files and scripts (Python) 166-way multi-species alignment file underlying FlyCADD (.maf file) Reconstructed ancestral sequence underlying FlyCADD (.fasta files) Derived and simulated variants used for model training and testing FlyCADD (.vcf files) FlyCADD scores at all codon positions of unique transcripts in D. melanogaster (.txt files) Additional resources The full pipeline of FlyCADD development is available on GitHub (https://github.com/JuliaBeets/FlyCADD/). For any additional information regarding the collection and generation of data please contact us.### 研究目的 单核苷酸多态性（Single Nucleotide Polymorphisms, SNPs）作为基因组变异最常见的形式，在微进化与适应过程中发挥关键作用。在黑腹果蝇（Drosophila melanogaster）中，诸多单核苷酸多态性已通过关联研究与表型建立关联，但功能验证仍颇具挑战，且支持其功能影响的实验证据较为稀缺。 本研究推出FlyCADD——一款将高质量黑腹果蝇基因组注释整合为单一评分的变异影响预测工具，该评分可反映单核苷酸多态性的预测影响程度。FlyCADD可用于区分致病变异与中性变异，具体应用场景包括：1）对单核苷酸多态性进行排序与优先级划分，以助力功能研究；2）优化基因组编辑实验的设计与评估流程；3）增强对天然存在的单核苷酸多态性的解读能力，从而加深我们对基因型-表型关联的理解。 ### 数据集内容 本数据集提供便捷可用的FlyCADD变异影响预测评分，既包含黑腹果蝇参考基因组上所有潜在单核苷酸变异的预计算评分，也提供可本地执行的分析流程，用于对目标新变异进行评分。若您需要获取目标单核苷酸多态性对应的FlyCADD评分，通常可直接在预计算的FlyCADD评分文件中查找，无需运行本地流程。如有任何疑问，欢迎联系[e-mail]。 本代码仓库包含以下内容： 1. 黑腹果蝇参考基因组Release 6版本上所有潜在单核苷酸多态性的预计算FlyCADD评分（.csv格式文件） 2. 用于对新变异进行评分的本地可执行FlyCADD分析流程，包含训练完成的逻辑回归模型文件、注释文件及Python脚本 3. FlyCADD所依赖的166物种多序列比对文件（.maf格式文件） 4. FlyCADD所依赖的重建祖先序列文件（.fasta格式文件） 5. 用于FlyCADD模型训练与测试的衍生及模拟变异文件（.vcf格式文件） 6. 黑腹果蝇唯一转录本所有密码子位置对应的FlyCADD评分（.txt格式文件） ### 附加资源 FlyCADD开发的完整流程已托管至GitHub（https://github.com/JuliaBeets/FlyCADD/）。若需了解数据收集与生成的更多细节，欢迎联系我们。