使用GATKA中的 IndexFeatureFile 工具来构建索引。gatk --java-options "-Xmx50g" IndexFeatureFile -I **.g.vcf.gz
成功得到了gvcf文件后，合并gvcf文件时，在导入db步骤中，由于gvcf文件超出了idx索引的处理范围，因此我使用CSI建立索引后，发现导入db的代码依旧无法运行，需要idx格式的索引，怎么办
2 个回答
擅长：重测序,遗传进化,转录组,GWAS
The problem is in the zipped (g.vcf.gz) files, for big genomes (chromosomes with size of > 500 mbp), the tbi index format can handle chromosomes up to ~ 530 Mbp. Indexing with csi is the option for large genomes with large chromosomes, but unfortunately, CombineGVCF and GenomicsDB don't accept it.
The only solution I found, is to work with unzipped g.vcf files (g.vcf)
Indexing with tabix is mandatory for gzipped vcf files (g.vcf.gz).
If files are unzipped, tbi indexing is not required. And it seems that GenomicsDB and combineGVCFs work ok with unzipped gVCF files.
Of course, the .idx is necessary. But .tbi is not required for unzipped vcf files.
参考这里：https://gatk.broadinstitute.org/hc/en-us/community/posts/4407400443803-GenomicsDBimport-and-CombineGVCF-does-not-show-variants-at-500-Mbp-onwards-although-gvcf-files-from-HapolypeCaller-report-variants
超过500M的染色体，GATK 索引不支持；
1.解压g.vcf.gz
gunzip demo.g.vcf.gz
2. 解压后的vcf文件建立索引：
gatk --java-options "-Xmx50g" IndexFeatureFile -I demo.g.vcf3.用解压后的g.vcf导入数据库如果觉得我的回答对您有用，请随意打赏。你的支持将鼓励我继续创作！
