Probe signal_Classification.rar

<br>Probe signals_ClassificationMachine learning was performed by MATLAB. More than 400 events of each analyte type were collected to form a dataset. The label for each event was assigned with the known identity of the analyte. The dataset was then split into a training set (80%) and a testing set (20%) for model training and model testing. I/I0 and SD of events were employed as event features. Model training was performed using the Classification Learner toolbox of MATLAB. Mainstream classifiers were estimated with default settings. According to results of five cross validation accuracy and the testing accuracy, the model demonstrating the best performance would be chosen for further use. ****************************************************************************************************************************************************1. System requirements Hardware Requirements: RAM: 16+ GB CPU: 4+ cores, 3.3+ GHz/core GPU: NVIDIA GTX 1080 Software Requirements: MATLAB R2022b2. Introduction Probe signals_Classification is meant to serve as a machine learning-based nanopore analysis platform using MATLAB for the identification of different PTMs peptides. The program contains two steps as follows: a. Import dataset. b. Open "Classification Learner" APP in MATLAB. c. Train Classifier splits the dataset into a training dataset(80%) and a testing dataset(20%), then outputs training accuracy, testing accuracy and confusion matrix. 3. Operating procedures: <br> -Unzip the Probe signals_Classification.rar to local folder. <br> -Open either Probe signals_Classification.m in MATLAB<br> -Enter the file name of the training dataset. // example: dataset.xlsx<br> -Run Probe signals_Classification.m<br>4. Output The values of the variables 'validationAccuracy' and 'TestingAccuracy ' are output in the MATLAB.