Amino acid-classifier

Amino acid_Classification

****************************************************************************************************************************************************

1. Introduction

Amino acid_Classification is an automatic classification system of amino acid events based on machine learning.

This code contains four steps as follows:

a. Import data.

b. Train classifier.

c. Make predictions with the returned "trainedClassifier" on predicting data.

d. Evaluate the classifier with the learning curve.

2. Operating procedures:

-Unzip the Amino acid-Classifier.rar to local folder.

-Open amino_acid_classifier.m in MATLAB.

-Enter the file name of the training set in line 4. // example: training_set_20aa.xlsx

-Enter the file name of the testing set in line 5. // example: testing_set_20aa.xlsx

-Enter the file name of the predicting dataset in line 6. // example: predicting_set_20aa.xlsx

-Run amino_acid_classifier.m

3. Output

a. The values of the variables 'validationAccuracy' and 'testingAccuracy ' are output in the command window, they are respectively the ten-fold cross-validation accuracy and the test accuracy of the classifier.

b. The confusion matrix, scatter plot with predicted labels and learning curve are output in the figure panel and saved as 'output.jpg' in local folder.