The Platform Development For The Analysis Of Autophagic Phenotypes And Database Construction

Autophagy is a highly conserved process in eukaryotes for the turnover of intracellular substances.The autophagic process is dynamically orchestrated by protein products of AuTophaGy-related(ATG)genes and autophagy regulators to maintain the metabolic balance between synthesis,degradation and recycling of cellular materials or organelles.As an important metabolic pathway to maintain cell homostasis,autophagy has been increasingly recognized to undergo rhythmic variation in accordance with circadian patterns of rest/activity and feeding in mammals.Both autophagy and circadian dysfunction have been implicated in the pathogenesis of aging,neurodegenerative disorder and cancer,suggesting a possible role of circadian autophagy to eliminate damaged substances and organelles from cells at regular intervals and maintain optimal physiological status.In recent years,benefited from the positive genetics to characterize phenotype and the negative genetics to study molecular mechanisms,a large number of experimental data have been identified to regulate autophagy and circadian rhythm.The rapid development of high-throughput technology also promoted the large-scale identification of transcription factors and post-translational modifications(PTMs)involved in autophagy and circadian rhythm.It has become an urgent task to collect these data and systematically analyze the regulation between autophagy and circadian rhythm,as well as establishing a fast and effective functional screening platform to facilitate experimental research.In this work,we developed an autophagic phenotype prediction tool based on fluorescent microscopy.Moreover,we performed a systematic bioinformatics analysis of autophagy and circadian genes.With an easy to manipulate genetic background and a highly conserved autophagy process,yeast has become the most well-studied model organism for autophagy.Although numerous methods have been developed to monitor autopahgy in yeast,there is still no one suitable for large-scale screening of regulatory proteins.GFP-Atg8 fluorescence assay can be used as a relatively accurate measure of cell autophagy activity,but yet to be widely adopted for functional screenings,due to the labor-intensive,time-consuming and error-prone process of manually labeling cells.In this work,we developed a novel software package named DeepPhagy for autophagic phenotype prediction and quantitative analysis.The performance of DeepPhagy was critically evaluated and compared with other existing tools,immunoblotting assays were also conducted and compared with computational recognitions,indicating DeepPhagy a useful tool for measuring autophagy activity.Finally,we systematically analyzed the protein-protein interactions(PPIs)of Atg proteins and found that all Atg proteins with strong phenotypic defect are involved in autophagosome formation and highly conserved in H.sapiens.Circadian rhythms are fundamental phenomena of life that are manifested as daily oscillations in vast biological processes and driven by endogenous clocks.From the simplest single cell cyanobacteria to plants,invertebrates and,at the highest levels,humans,the molecular basis of circadian rhythms lies in a set of transcription–translation feedback loops(TTFLs)that drives the rhythmic transcription of core clock genes.In this work,1,382 circadian genes identified by small-scale experiments have been collected from literature manually,some of these oscillating genes are part of the clock or directly controlled by the clock.We also collected 26,582 oscillating genes that were identified by microarray and/or RNA-seq,which might be indirectly regulated by the clock or driven by rhythmic changes in the environment.In addtion,we conducted an orthologous search and further identified 44,836 potential oscillating genes in 148 eukaryotes.For each circadian gene,the oscillation information regarding the external condition,peak and trough time points of the oscillation,the amplitude values,as well as the tissue/cells have been curated.Finally,we developed a comprehensive circadian gene database(CGDB 1.0)(http://cgdb.biocuckoo.org),containing 72,800 non-redundant cycling genes integrated with oscillation information,post translation modifications,reference literature,ortholog information,et al.Based on the above results,we curated experimentally identified genes involved in autophagy which were further integrated into THANATOS(http://thanatos.biocuckoo.org/).In order to systematically analyze the regulation of circadian and autophagy,we extracted autophagy genes which exihibit circadian rhythm(circadian autophagy genes)and analyzed the function of circadian autophagy genes in mouse and human.Finally,we visualized the autophagy regulating and circadian rhythm pattern of circadian autophagy genes in “autophagy regulating pathway” from mouse.Taken together,we applied deep learning in the analysis of autophagic phenotypes from fluorescent images,making it possible for large-scale functional screenings.Meanwhile,we curated genes involved in circadian rhythm and autophagy and integrated them into CGDB and THANATOS.Furthermore,we systematically analyze and visualized the function of circadian autophagy genes,providing novel research ideas for further analyzing the regulation mechanisms of autophagy and circadian rhythm.