Background: Genome Reference Consortium Human Build 38 (GRCh38) was released with improvements, such as accuracy and completeness, over human genome 19 (hg19). However, GRCh38 has not been widely adopted because realignment is time-consuming and computationally expensive. To address this issue, faster and more convenient liftover tools have been developed to convert genome coordinates between assemblies. This study investigated the differences in genetic variants detected according to reference genome selection. Additionally, we investigated the accuracy of liftover tools for the conversion to GRCh38 from hg19.
Methods: Four FASTQ files (GNG-22-01, GNG-22-02, GNG-22-04, and GNG-22-05) and validated variant descriptions (n=144) were provided by the Korean Association of External Quality Assessment Service. The variants detected based on the alignment to hg19 were compared to those based on the alignment to GRCh38. The liftover tools, such as CrossMap, NCBIRemap, and UCSCliftOver, were used to convert the genome coordinates from hg19 to GRCh38.
Results: Among the variants identified based on the hg19 alignment, 2% (3/144) were not detected based on the GRCh38 alignment: NM_000219.6 (KCNE1): c.112A>G (p.Ser38Gly), NM_005429.5(VEGFC): c.1256_1258delTGA (p.Met419del), NM_004415.4(DSP):c.741T>G (p.Ala247=). The accuracy of CrossMap, NCBIRemap, and UCSCliftOver was 100% (10,725/10,725), 99.99% (10,724/10,725), and 100% (10,725/10,725), respectively.
Conclusions: This study suggests that liftover tools might be one of the practical alternatives for genome conversion in cases where realignment approaches are not possible. Further clinical studies are warranted to compare the performance of liftover tools and realignment approaches.

Keywords: Alignment, Liftover, Genome Reference Consortium Human Build 38, Human genome 19, CrossMap, NCBIRemap, UCSCliftOver