Despite great efforts by experimental biologists, biological data enter into the databanks at far greater pace with the fast advances of high-throughout screening technologies in the post-genomic Era [1-3]. We are facing the tough fact that huge gap exists between newly found data and their knowledge. Fruitful bioinformatics research results have made computational models be widely accepted as one of the major solutions to deal with the mass biological data . The most successful current bioinformatics tools are built utilizing modern advanced intelligent modeling technologies, including statistical machining learning algorithms, computational graph theory, and sampling theory etc [5-12]. Monitoring the state-of-the-art development of this dynamic and quickly advancing field is of critical importance to keep the users informed so that they can make the best use of these computational models. This issue provides a comprehensive overview in this regard, which consists of two parts, the first with four reviews and the second including four research papers.
The first review by Zheng et.al summarizes the recent advancement of methodology development in characterizing and predicting human genomic DNA methylation, which is a major epigenetic modification that adds a methyl group mainly to the carbon-5 position of the cytosine pyrimidine ring in the cytosine guanine dinucleotide and is crucial to normal development and cellular differentiation. The second paper by Xu et.al reviews the development of identifying coevolution between amino acid residues in protein families, and then further discusses the inconsistent results generated by different algorithms and their possible reasons as well as future challenges. The review by Wang et.al discusses the computational methods and available tools in predicting functional impact of single amino acid polymorphisms. The last review introduces the statistical machine learning-based approaches for predicting protein-protein interaction sites.
The first research article by Fan et.al proposes a two-layer framework for predicting protein N-terminal signal peptide and their cleavage sites, where the first layer is achieved by hydrophobicity alignment and position-specific amino acid propensities, and the second layer is constructed with conditional random filed algorithm. In the next paper, Du et.al introduces their method SubChlo-GO for predicting protein subchloroplast locations with weighted gene ontology scores. The work by Chen et.al describes a method based on the combination of graph property, chemical functional group, and chemical structure set for predicting metabolic pathway. We close the issue with the paper by Paci et.al that describes the subcellular and stem cell image classification protocol based on non-binary coding for texture descriptors.
We are excited to deliver this special issue that deals with different areas in bioinformatics. We hope that it will be useful for the scientists of bioinformatics, wet-lab experiments, and computer science. Last but not least, we would like to thank all the authors and anonymous reviewers who make this issue possible. We are also in great debt to the Editor-in-Chief Prof. Satya P. Gupta for his invitation and support that resulted in the successful completion of this special issue.