Journal
FRONTIERS IN PLANT SCIENCE
Volume 9, Issue -, Pages -Publisher
FRONTIERS MEDIA SA
DOI: 10.3389/fpls.2018.00634
Keywords
machine learning; mitochondrial targeting; deep neural network; support vector machine; position weight matrix; gene co-expression
Categories
Funding
- National Institutes of Health [R01-GM100701]
- National Science Foundation [DBI-1126992, CNS-1429294]
- National Institutes of Health
- Danish Council for Independent Research Natural Sciences (FNU)
- Division Of Computer and Network Systems
- Direct For Computer & Info Scie & Enginr [1429294] Funding Source: National Science Foundation
Ask authors/readers for more resources
Targeting and translocation of proteins to the appropriate subcellular compartments are crucial for cell organization and function. Newly synthesized proteins are transported to mitochondria with the assistance of complex targeting sequences containing either an N-terminal pre-sequence or a multitude of internal signals. Compared with experimental approaches, computational predictions provide an efficient way to infer subcellular localization of a protein. However, it is still challenging to predict plant mitochondrially localized proteins accurately due to various limitations. Consequently, the performance of current tools can be improved with new data and new machine-learning methods. We present MU-LOC, a novel computational approach for large-scale prediction of plant mitochondrial proteins. We collected a comprehensive dataset of plant subcellular localization, extracted features including amino acid composition, protein position weight matrix, and gene co-expression information, and trained predictors using deep neural network and support vector machine. Benchmarked on two independent datasets, MU-LOC achieved substantial improvements over six state-of-the-art tools for plant mitochondrial targeting prediction. In addition, MU-LOC has the advantage of predicting plant mitochondrial proteins either possessing or lacking N-terminal pre-sequences. We applied MU-LOC to predict candidate mitochondrial proteins for the whole proteome of Arabidopsis and potato. MU-LOC is publicly available at http://mu-loc.org.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available