Machine learning provides an important technical support for many cross disciplines. For example, "bioinformatics" attempts to use information technology to study life phenomena and laws, while the implementation of the genome project and the bright prospect of gene drugs make people excited about it. Bioinformatics research involves the whole process from "life phenomena" to "law discovery", which must include data acquisition, data management, data analysis, simulation experiments and other links. "Data analysis" is just the stage of machine learning technology, and various machine learning technologies have been brilliant on this stage. The expanding scale and inherent complexity of biological data promote the application of machine learning in biology more and more. All machine learning technologies can match models with data. However, there are still many puzzles about how to correctly understand and use machine learning technology for biological researchers.
In order to enable more teams to use neural networks as a tool to complete some modeling work, our team has developed a single variable output BP neural network prediction software. This software was developed using the Guide function of Matlab. Users only need to import the training set, test set and data set to be predicted according to the requirements of the software, and then simply adjust the parameters related to the neural network in the software, such as the number of nodes, learning rate, error requirements, neural network training algorithm, etc. After the user has trained the neural network, our software will draw the fitting curve and calculate the goodness of fit and error. Users can also export the trained network as Mat file which is convenient for users to use neural network on Matlab later. In the process of software development, we found that the Guide function will be gradually replaced in the versions after R2021 of Matlab. Considering this reason, in order to increase the universality of our software, we re developed the software with the App Designer function of the newer version of Matlab. There is no significant difference between the two versions of the software in terms of usage. The running interface of the software is as follows:
Software Introduction Video