「GATK 3.8」FastaStats和CountReads介绍

GATK的本职工作是Variant calling，但是就像我之前所说的，它作为基因组分析工具箱，还是有很多其他工具，今天学习的是诊断和质量控制工具的其中两个:CountReads，FastaStats。

FastaStats

功能： 计算参考基因组的基本统计值
分类： 诊断和质量控制工具
概要： 主要就是统计碱基总数，和常规的碱基数（ATCG）
吐槽：功能还真是简单。。

输入： FASTA参考文件
输出： 结果要么输出到屏幕，要么是输出到(-o)到文件中。
使用案例：我看了一下拟南芥的参考基因组

java -jar ~/biosoft/GenomeAnalysisTK.jar -T FastaStats \
    -R TAIR10.fa
输出：
Total bases   119667750
Regular bases 119481543

CountReads

功能： 计算reads数
分类： 诊断和质量控制工具
概要： 最好和--read-filter合用，这样可以了解下符合特定标准的reads数
输入： 一个或多个BAM文件
输出： 结果会输出到屏幕（标准输出）上，毕竟是用来确定阈值的，也不需要一定要输出到文件中

不加参数--read-filter

java -jar ~/biosoft/GenomeAnalysisTK.jar -T CountReads -R $work/database/TAIR10/TAIR10.fa  -I BC_bg_reads.sorted.bam
输出：
CountReads - CountReads counted 55080781 reads in the traversal

--read-filter/-rf后面可以接很多的选项，官方文档列出了如下内容：

Name	Summary
BadCigarFilter	Filter out reads with wonky CIGAR strings
BadMateFilter	Filter out reads whose mate maps to a different contig
CountingFilteringIterator.CountingReadFilter
DuplicateReadFilter	Filter out duplicate reads
FailsVendorQualityCheckFilter	Filter out reads that fail the vendor quality check
HCMappingQualityFilter	Filter out reads with low mapping qualities for HaplotypeCaller
LibraryReadFilter	Only use reads from the specified library
MalformedReadFilter	Filter out malformed reads
MappingQualityFilter	Filter out reads with low mapping qualities
MappingQualityUnavailableFilter	Filter out reads with no mapping quality information
MappingQualityZeroFilter	Filter out reads with mapping quality zero
MateSameStrandFilter	Filter out reads with bad pairing (and related) properties
MaxInsertSizeFilter	Filter out reads that exceed a given insert size
MissingReadGroupFilter	Filter out reads without read group information
NoOriginalQualityScoresFilter	Filter out reads that do not have an original quality quality score (OQ) tag
NotPrimaryAlignmentFilter	Filter out read records that are secondary alignments
OverclippedReadFilter	Filter out reads that are over-soft-clipped
Platform454Filter	Filter out reads produced by 454 technology
PlatformFilter	Filter out reads that were generated by a specific sequencing platform
PlatformUnitFilter	Filter out reads with blacklisted platform unit tags
ReadGroupBlackListFilter	Filter out reads matching a read group tag value
ReadLengthFilter	Filter out reads based on length
ReadNameFilter	Only use reads with this read name
ReadStrandFilter	Filter out reads based on strand orientation
ReassignMappingQualityFilter	Set the mapping quality of all reads to a given value
ReassignOneMappingQualityFilter	Set the mapping quality of reads with a given value to another given value
ReassignOriginalMQAfterIndelRealignmentFilter	Revert the MQ of reads that were modified by IndelRealigner
SampleFilter	Only use reads belonging to a specific sample
SingleReadGroupFilter	Only use reads from the specified read group
UnmappedReadFilter	Filter out unmapped reads

使用的时候记住，对于XXXXFilter，你需要点每个过滤选项进去看下具体用法，比如说我需要过滤比对质量低于20的,我发现MappingQualityFilter的用法解释是

  java -jar GenomeAnalysisTk.jar \
         -T HaplotypeCaller \
         -R reference.fasta \
         -I input.bam \
         -o output.vcf \
         -rf MappingQuality \
         -mmq 15

那么我前面的代码就可以改为

java -jar ~/biosoft/GenomeAnalysisTK.jar -T CountReads -R $work/database/TAIR10/TAIR10.fa  -I BC_bg_reads.sorted.bam \
-rf MappingQuality -mmq 15

结果：
8190205 reads were filtered out during the traversal out of approximately 55080781 total reads (14.87%)

果然加上比对质量后，就有一些要被过滤掉呢。

基本上GATK每一个工具都有一些默认过滤选项，你可以在每个工具的额外信息部分进行了解。

今天就介绍这两个工具，顺便提了GATK的过滤功能。白了个白，每天学个GATK的任务完成。