The artical introduces a new algorithm that combines hash table and suffix tree techniques to significantly speed up the alignment of short-read NGS data against large reference genomes like the human genome. This algorithm reduces the processing time of a human genome sample from over 25 hours to just 31 minutes, achieving a 4.4-fold increase in speed while maintaining comparable accuracy to traditional methods. It also supports fast, efficient variant calling with existing open-source tools and a proprietary variant caller.
