Abstract
The wide functional impacts of microtubules are unleashed and controlled by a battery of microtubule-associated proteins (MAPs). Specialists in the field appreciate the diversity of known MAPs and propel the identifications of novel MAPs. By contrast, there is neither specific database to record known MAPs, nor MAP predictor that can facilitate the discovery of potential MAPs. We here report the establishment of a MAP-centered online analysis tool MAPanalyzer, which consists of a MAP database and a MAP predictor. In the database, a core MAP dataset, which is fully manually curated from the literature, is further enriched by MAP information collected via automated pipeline. The core dataset, on the other hand, enables the building of a novel MAP predictor which combines specialized machine learning classifiers and the BLAST homology searching tool. Benchmarks on the curated testing dataset and the Arabidopsis thaliana whole genome dataset have shown that the proposed predictor outperforms not only its own components (i.e. the machine learning classifiers and BLAST), but also another popular homology searching tool, PSI-BLAST. Therefore, MAPanalyzer will serve as a promising computational resource for the investigations of MAPs. Database URL: http://systbio.cau.edu.cn/mappred/.