Performance assessment of variant calling pipelines using human whole exome sequencing and simulated data

Background

Whole exome sequencing (WES) is a cost-effective method that identifies clinical variants but it demands accurate variant caller tools. Currently available tools have variable accuracy in predicting specific clinical variants. But it may be possible to find the best combination of aligner-variant caller tools for detecting accurate single nucleotide variants (SNVs) and small insertion and deletion (InDels) separately. Moreover, many important aspects of InDel detection are overlooked while comparing the performance of tools, particularly its base pair length.

Results

We assessed the performance of variant calling pipelines using the combinations of four variant callers and five aligners on human NA12878 and simulated exome data. We used high confidence variant calls from Genome in a Bottle (GiaB) consortium for validation, and GRCh37 and GRCh38 as the human reference genome. Based on the performance metrics, both BWA and Novoalign aligners performed better with DeepVariant and SAMtools callers for detecting SNVs, and with DeepVariant and GATK for InDels. Furthermore, we obtained similar results on human NA24385 and NA24631 exome data from GiaB.

Conclusion

In this study, DeepVariant with BWA and Novoalign performed best for detecting accurate SNVs and InDels. The accuracy of variant calling was improved by merging the top performing pipelines. The results of our study provide useful recommendations for analysis of WES data in clinical genomics.

Electronic supplementary material

The online version of this article (10.1186/s12859-019-2928-9) contains supplementary material, which is available to authorized users.

hg19KIndel: ethnicity normalized human reference genome

Background

The most widely used human genome reference assembly hg19 harbors minor alleles at 2.18 million positions as revealed by 1000 Genome Phase 3 dataset. Although this is less than 2% of the 89 million variants reported, it has been shown that the minor alleles can result in 30% false positives in individual genomes, thus misleading and burdening downstream interpretation. More alarming is the fact that, significant percentage of variants that are homozygous recessive for these minor alleles, with potential disease implications, are masked from reporting.

Results

We have demonstrated that the false positives (FP) and false negatives (FN) can be corrected for by simply replacing nucleotides at the minor allele positions in hg19 with corresponding major allele. Here, we have effectively replaced 2.18 million minor alleles Single Nucleotide Polymorphism (SNPs), Insertion and Deletions (INDELs), Multiple Nucleotide Polymorphism (MNPs) in hg19 with the corresponding major alleles to create an ethnically normalized reference genome called hg19KIndel. In doing so, hg19KIndel has both corrected for sequencing errors acknowledged to be present in hg19 and has improved read alignment near the minor alleles in hg19.

Conclusion

We have created and made available a new version human reference genome called hg19KIndel. It has been shown that variant calling using hg19KIndel, significantly reduces false positives calls, which in-turn reduces the burden from downstream analysis and validation. It also improved false negative variants call, which means that the variants which were getting missed due to the presence of minor alleles in hg19, will now be called using hg19KIndel. Using hg19KIndel, one even gets a better mapping percentage when compared to currently available human reference genome. hg19KIndel reference genome and its auxiliary datasets are available at 10.5281/zenodo.2638113

Improvements and impacts of GRCh38 human reference on high throughput sequencing data analysis

Analyses of high throughput sequencing data starts with alignment against a reference genome, which is the foundation for all re-sequencing data analyses. Each new release of the human reference genome has been augmented with improved accuracy and completeness. It is presumed that the latest release of human reference genome, GRCh38 will contribute more to high throughput sequencing data analysis by providing more accuracy. But the amount of improvement has not yet been quantified. We conducted a study to compare the genomic analysis results between the GRCh38 reference and its predecessor GRCh37. Through analyses of alignment, single nucleotide polymorphisms, small insertion/deletions, copy number and structural variants, we show that GRCh38 offers overall more accurate analysis of human sequencing data. More importantly, GRCh38 produced fewer false positive structural variants. In conclusion, GRCh38 is an improvement over GRCh37 not only from the genome assembly aspect, but also yields more reliable genomic analysis results.