A Markov chain-based feature extraction method for classification and identification of cancerous DNA sequences

Abstract

Introduction: In recent decades, the growing rate of cancer incidence is a big concern for most societies. Due to the genetic origins of cancer disease, its internal structure is necessary for the study of this disease. Methods: In this research, cancer data are analyzed based on DNA sequences. The transition probability of occurring two pairs of nucleotides in DNA sequences has Markovian property. This property inspires the idea of feature dimension reduction of DNA sequence for overcoming the high computational overhead of genes analysis. This idea is utilized in this research based on the Markovian property of DNA sequences. This mapping decreases feature dimensions and conserves basic properties for discrimination of cancerous and non-cancerous genes. Results: The results showed that a non-linear support vector machine (SVM) classifier with RBF and polynomial kernel functions can discriminate selected cancerous samples from non-cancerous ones. Experimental results based on the 10-fold cross-validation and accuracy metrics verified that the proposed method has low computational overhead and high accuracy. Conclusion: The proposed algorithm was successfully tested on related research case studies. In general, a combination of proposed Markovian-based feature reduction and non-linear SVM classifier can be considered as one of the best methods for discrimination of cancerous and non-cancerous genes.

A Robust Feature Extraction and Deep Learning Approach for Cancer Gene Prognosis

“…The dataset is an imbalanced dataset with 82% cancerous genes and 18% noncancerous genes. Existing model Khodaei et al [6] is a dataset of one input feature. Compared with the existing models, the accuracy of our cancer gene prediction model is increased by 1.2%.…”

“…Step 5. The desired output is evaluated from the below mentioned input output relation (6) Step 6 : Estimated error is evaluated as (7)…”

“…Khodaei et al [6] proposed a hybrid approach to introduce machine learning into cancer prediction. In their work, Markov chain model is deployed for feature extraction and SVM as classifier.…”

Application of Artificial Intelligence in Diagnosis of Craniopharyngioma

“…The general evaluation criteria include Fisher score ( 58 ), Pearson correlation coefficient ( 59 ), mutual information ( 60 ), etc. In addition, some heuristic rules can also be applied in selecting subsets, such as forward/backward search strategy ( 61 ), Markov chain ( 62 ), etc. It can also combine the criterion-based sorting method with the search strategy to form a two-step feature selection method, such as the feature selection method with maximum relation and minimum redundancy ( 63 ), which obtained good result in reports.…”