Taguchi, Y.-h.; Turki, T.

Genome-wide CRISPR-Cas9 knockout screens is often used to evaluate gene functions experimentally. Nevertheless, the efficacy of individual sgRNAs that target unique gene varies and is hard to integrate. In this study, tensor decomposition (TD) was used to integrate multiple sgRNAs as well as sgRNA profiles simultaneously. As a result, TD can discriminate between essential and non-essential genes with the competitive performances of one of SOTA, JACKS, which previously outperformed various SOTA. In addition to this, in spite of that TD is nothing but the simple linear algebra, TD can achieve the performance even without control samples without which JACKS even cannot be performed. Moreover, since raw and logarithmic values can achieve the similar performances through TD for the largest dataset among those tested, taking logarithmic values as has been done frequently is questioned. In conclusion, TD is the ever first method that can integrate multiple sgRNAs and profiles simultaneously and can achieve comparative performances with JACKS, one of SOTA.