Groovy NGS provides a dedicated VCF parser designed to offer highly streamlined access to VCF information, via the https://ssadedin.github.io/groovy-ngs-utils/doc/gngs/VCF.html gngs.VCF class. This parser allows access to key header properties as well as full information about variants.To load a VCF into memory, use the https://ssadedin.github.io/groovy-ngs-utils/doc/gngs/VCF.html#parse(Closure) VCF.parse static method. This method accepts a closure which, if provided, is called for each variant to determine if the variant should be retained. This allows loading a subset of variants from a VCF. For example, here we load only variants on chromosome 22. Input groovy vcf = VCF.parse("test.vcf") { it.chr == 'chr22' } println "The samples are: $vcf.samples" println "There are ${vcf.size()} variants on chr22" Output text The samples are: [NA12877, NA12878, NA12879] There are 24 variants on chr22

If access to the variant information is not required, you can avoid parsing the whole VCF by calling the constructor directly: Input groovy vcf = new VCF("test.vcf") println "The samples are: $vcf.samples" Output text The samples are: [NA12877, NA12878, NA12879]

For large VCFs it can be beneficial to process a VCF file line by line. In this case, the https://ssadedin.github.io/groovy-ngs-utils/doc/gngs/VCF.html#filter(java.lang.String,%20Closure) VCF.filter method can be used to process variants in streaming fashion. By default, the filter method reads from standard input and writes to standard output, but a file names can be provided to read from a file. groovy VCF.filter('test.vcf') { it.chr == 'chr22' // output only variants on chr22 to stdout }

Each variant is represented by an instance of the https://ssadedin.github.io/groovy-ngs-utils/doc/gngs/Variant.html gngs.Variant class which makes available detailed information about the variant. The VCF class implements the Iterable<Variant> interface, making available standard operations that are available on all Groovy https://docs.groovy-lang.org/latest/html/groovy-jdk/java/util/Iterator.html Iterators. Input groovy vcf = VCF.parse("test.vcf").countBy { it.type } Output text {SNP=22, DEL=1, INS=1}

When accessing a small number of variants from a large VCF, it is more efficient to index the VCF and parse only the relevant lines. The https://ssadedin.github.io/groovy-ngs-utils/doc/gngs/VCFIndex.html gngs.VCFIndex class can be used to query variants within a specified genomic region. Input groovy index = new VCFIndex("test.vcf.gz") index.iterator("chr1", 13792200, 13792300).collect { it } Output text [chr1:13792284 G/T]