Deep Forest-based Prediction of Protein Subcellular Localization

摘要

Motivation: Knowledge of the correct protein subcellular localization is necessary for understandingthe function of a protein and revealing the mechanism of many human diseases due to proteinsubcellular mislocalization, which is required before approaching gene therapy to treat a disease. In addition,it is well-known that the gene therapy is an effective way to overcome disease by targeting a genetherapy product to a specific subcellular compartment. Deep neural networks to predict protein functionhave become increasingly popular due to large increases in the available genomics data due to itsstrong superiority in the non-linear classification ability. However, they still have some drawbackssuch as too many hyper-parameters and sufficient amount of labeled data.Results: We present a deep forest-based protein location algorithm relying on sequence information.The prediction model uses a random forest network with a multi-layered structure to identify thesubcellular regions of protein. The model was trained and tested on a latest UniProt releases proteindataset, and we demonstrate that our deep forest predict the subcellular location of proteins given onlythe protein sequence with high accuracy, outperforming the current state-of-art algorithms.Meanwhile, unlike the deep neural networks, it has a significantly smaller number of parameters andis much easier to train.