High Content Cellular Analysis for Functional Screening of Novel Cell Cycle Related Genes

Functional screening of novel genes can be efficiently performed by detecting morphological changes of transfected cells. In this study, we developed automated cellular analysis software and found novel genes that induced mitotic phenotypes in transiently transfected HeLa cells. Systemic imaging errors common in high content microscopy were corrected by background subtraction and intensity normalization. Nuclear and chromatin objects were detected by tophat operations. Shape, intensity, and convex-hull features were extracted. Cell types were classified by several logistic regression formulas with 93.9% accuracy in average. The increased % mitotic cells were detected in 47 out of 571 transfected genes. Twenty genes showed more than two-fold increase in % mitotic cells confirmed by manual inspection. Among these, nine genes showed increased tissue expression levels in several tumors, thus, indicating their possible oncogenic roles. These genes will be further investigated biochemically to confirm their cell cycle related functions.