Autophagy is an evolutionarily conserved pathway from yeast to human for degrading and recycling cellular components and exerts in both bulk/nonselective and selective manners (Reggiori et al., 2002). The autophagic process is dynamically orchestrated by protein products of AuTophaGy-related (ATG) genes and autophagy regulators, where the dysfunction of autophagy has been implicated in a wide range of human diseases including cancer, diabetes and neurodegenerative disorders (Levine et al., 2011). Accurately monitoring and measuring autophagic activity are fundamental for understanding the functional importance and regulatory roles of autophagy under physiological and pathological conditions (Elizabeth et al., 2015).

Here we present a novel software of DeepPhagy (Deeplearning for autoPhagy) as a starting point for the application of deep learning-based computer vision approaches in the analysis of autophagic phenotypes from fluorescent images. DeepPhagy was written in Java 1.8 (J2SE8) and packaged with Install4j 6.0. Thus, the DeepPhagy 1.0 can be easily installed on a computer. Then we developed several packages to support three major operating systems (OS), including Windows, Unix/Linux, and Mac. The Windows 7/10, Ubuntu, Apple Mac OS 10.12 (Sierra) were chosen to test the stability of DeepPhagy 1.0, a Java Runtime Environment 8 (JRE8) package of Oracle should be installed first or you can choose install package which include JRE.

DeepPhagy 1.0 User Interface

For publication of results please cite the following articles: Deepphagy: a deep learning framework for quantitatively measuring the autophagy activity in Saccharomyces cerevisiae Ying Zhang, Yubin Xie, Wenzhong Liu, Wankun Deng, Di Peng, Chenwei Wang, Haodong Xu, Chen Ruan, Yongjie Deng, Yaping Guo, Cong Yi, Jian Ren and Yu Xue.