pLoc_Deep mHum: Predict Subcellular Localization of Human Proteins by Deep Learning

Recently, the life of human beings around\r\nthe entire world has been endangering by the spreading of pneumonia-causing\r\nvirus, such as Coronavirus, COVID-19, and H1N1. To develop effective drugs\r\nagainst Coronavirus, knowledge of protein subcellular localization is\r\nindispensable. In 2019, a predictor called “pLoc_bal-mHum” was developed for\r\nidentifying the subcellular localization of human proteins. Its predicted\r\nresults are significantly better than its counterparts, particularly for those\r\nproteins that may simultaneously occur or move between two or more subcellular\r\nlocation sites. However, more efforts are definitely needed to further improve\r\nits power since pLoc_bal-mHum was still not trained by a “deep learning”, a\r\nvery powerful technique developed recently. The present study was devoted to\r\nincorporate the “deep-learning” technique and develop a new predictor called\r\n“pLoc_Deep-mHum”. The global absolute true rate achieved by the new predictor\r\nis over 81% and its local accuracy is over 90%. Both are overwhelmingly\r\nsuperior to its counterparts. Moreover, a user-friendly web-server for the new\r\npredictor has been well established at\r\nhttp://www.jci-bioinfo.cn/pLoc_Deep-mHum/, which will become a very useful tool\r\nfor fighting pandemic coronavirus and save the mankind of this planet.