Perspectives of Bioinformatics in Big Data Era

Feature extraction approaches for biological sequences: a comparative study of mathematical features

As consequence of the various genomic sequencing projects, an increasing volume of biological sequence data is being produced. Although machine learning algorithms have been successfully applied to a large number of genomic sequence-related problems, the results are largely affected by the type and number of features extracted. This effect has motivated new algorithms and pipeline proposals, mainly involving feature extraction problems, in which extracting significant discriminatory information from a biological set is challenging. Considering this, our work proposes a new study of feature extraction approaches based on mathematical features (numerical mapping with Fourier, entropy and complex networks). As a case study, we analyze long non-coding RNA sequences. Moreover, we separated this work into three studies. First, we assessed our proposal with the most addressed problem in our review, e.g. lncRNA and mRNA; second, we also validate the mathematical features in different classification problems, to predict the class of lncRNA, e.g. circular RNAs sequences; third, we analyze its robustness in scenarios with imbalanced data. The experimental results demonstrated three main contributions: first, an in-depth study of several mathematical features; second, a new feature extraction pipeline; and third, its high performance and robustness for distinct RNA sequence classification. Availability:  https://github.com/Bonidia/FeatureExtraction_BiologicalSequences.

A Deep Convolutional Neural Network to Improve the Prediction of Protein Secondary Structure

Background:

Protein secondary structure prediction (PSSP) is a fundamental task in bioinformatics that is helpful for understanding the three-dimensional structure and biological function of proteins. Many neural network-based prediction methods have been developed for protein secondary structures. Deep learning and multiple features are two obvious means to improve prediction accuracy.

Objective:

To promote the development of PSSP, a deep convolutional neural network-based method is proposed to predict both the eight-state and three-state of protein secondary structure.

Methods:

In this model, sequence and evolutionary information of proteins are combined as multiple input features after preprocessing. A deep convolutional neural network with no pooling layer and connection layer is then constructed to predict the secondary structure of proteins. L2 regularization, batch normalization, and dropout techniques are employed to avoid over-fitting and obtain better prediction performance, and an improved cross-entropy is used as the loss function.

Results:

Our proposed model can obtain Q3 prediction results of 86.2%, 84.5%, 87.8%, and 84.7%, respectively, on CullPDB, CB513, CASP10 and CASP11 datasets, with corresponding Q8 prediction results of 74.1%, 70.5%, 74.9%, and 71.3%.

Conclusion:

We have proposed the DCNN-SS deep convolutional-network-based PSSP method, and experimental results show that DCNN-SS performs competitively with other methods.

Identifying Antioxidant Proteins by Using Amino Acid Composition and Protein-Protein Interactions

Excessive oxidative stress responses can threaten our health, and thus it is essential to produce antioxidant proteins to regulate the body’s oxidative responses. The low number of antioxidant proteins makes it difficult to extract their representative features. Our experimental method did not use structural information but instead studied antioxidant proteins from a sequenced perspective while focusing on the impact of data imbalance on sensitivity, thus greatly improving the model’s sensitivity for antioxidant protein recognition. We developed a method based on the Composition of k-spaced Amino Acid Pairs (CKSAAP) and the Conjoint Triad (CT) features derived from the amino acid composition and protein-protein interactions. SMOTE and the Max-Relevance-Max-Distance algorithm (MRMD) were utilized to unbalance the training data and select the optimal feature subset, respectively. The test set used 10-fold crossing validation and a random forest algorithm for classification according to the selected feature subset. The sensitivity was 0.792, the specificity was 0.808, and the average accuracy was 0.8.