Using Language Representation Learning Approach to Efficiently Identify Protein Complex Categories in Electron Transport Chain

• Published in: Molecular informatics 2020-10, Vol.39 (10), p.e2000033-n/a
• Main Authors: Nguyen, Trinh‐Trung‐Duong, Le, Nguyen‐Quoc‐Khanh, Ho, Quang‐Thai, Phan, Dinh‐Van, Ou, Yu‐Yen
• Format: Article
• Language: English
• Subjects: Algorithms; Amino acid sequence; Artificial neural networks; electron complexes; Electron transport; Electron transport chain; Embedding; Feature extraction; Language; Learning algorithms; Machine learning; Model accuracy; motif frequencies; Natural language processing; Neural networks; protein function prediction; Proteins; representation learning; Representations; Support vector machines; Transfer learning; word embeddings
• ISSN: 1868-1743; 1868-1751
• DOI: 10.1002/minf.202000033

We herein proposed a novel approach based on the language representation learning method to categorize electron complex proteins into 5 types. The idea is stemmed from the the shared characteristics of human language and protein sequence language, thus advanced natural language processing techniques were used for extracting useful features. Specifically, we employed transfer learning and word embedding techniques to analyze electron complex sequences and create efficient feature sets before using a support vector machine algorithm to classify them. During the 5‐fold cross‐validation processes, seven types of sequence‐based features were analyzed to find the optimal features. On an average, our final classification models achieved the accuracy, specificity, sensitivity, and MCC of 96 %, 96.1 %, 95.3 %, and 0.86, respectively on cross‐validation data. For the independent test data, those corresponding performance scores are 95.3 %, 92.6 %, 94 %, and 0.87. We concluded that using feature extracted using these representation learning methods, the prediction performance of simple machine learning algorithm is on par with existing deep neural network method on the task of categorizing electron complexes while enjoying a much faster way for feature generation. Furthermore, the results also showed that the combination of features learned from the representation learning methods and sequence motif counts helps yield better performance.