摘要
The Net Acet method has been developed to make predictions of N-terminalacetylation sites, but more information of the data set could be utilized to improve the performanceof the model. By employing a new way to extract patterns from sequences and using a samplebalancing mechanism, we obtained a correlation coefficient of 0.85, and a sensitivity of 93% on anindependent mammalian data set.
The Net Acet method has been developed to make predictions of N-terminalacetylation sites, but more information of the data set could be utilized to improve the performanceof the model. By employing a new way to extract patterns from sequences and using a samplebalancing mechanism, we obtained a correlation coefficient of 0.85, and a sensitivity of 93% on anindependent mammalian data set.
基金
This work was supported by the National Natural Science Foundation of China (No.10371063)the National Key Technologies R&D Program of China (No.2004ba711a21).
