Modeling positional effects of regulatory sequences with spline transformations increases prediction accuracy of deep neural networks.
Ziga AvsecMohammadamin BarekatainJun ChengJulien GagneurPublished in: Bioinform. (2018)
Keyphrases
- prediction accuracy
- neural network
- predictive power
- ensemble methods
- improve the prediction accuracy
- predictive accuracy
- pattern recognition
- genetic algorithm
- web page prediction
- artificial neural networks
- fuzzy logic
- b spline
- estimation accuracy
- improved accuracy
- least squares
- hidden markov models
- multiresolution
- feature space
- bayesian networks
- image segmentation