Abstract
This special issue of the Journal of Computational Biology includes a selection of articles presented at the 9th International Computational Advances in Bio and Medical Sciences (ICCABS 2019), which was held at Florida International University in Miami, Florida, during November 15–17, 2019. ICCABS has the goal of bringing together researchers, scientists, and students from academia, laboratories, and industry to discuss recent advances on computational techniques and applications in the areas of biology, medicine, and drug discovery.
In 2019, 30 extended abstracts were submitted in response to the ICCABS call for articles, out of which 14 extended abstracts appeared in the ICCABS 2019 postproceedings published as volume 12,029 of Springer Verlag's Lecture Notes in Bioinformatics series. Authors of seven articles were invited to submit extended versions of their abstracts to this special issue. In addition, authors of one talk presented at the 9th Workshop on Computational Advances for Next Generation Sequencing (CANGS 2019), jointly held with ICCABS 2019, were also invited to submit a full article to this special issue.
The first four articles use machine learning techniques to address computational problems in genomic data analysis, disease diagnosis, and treatment planning. Mohebbi et al. introduce a novel algorithm for predicting microRNA targets by simultaneously predicting the mechanism of targeting. Experimental results show improved prediction accuracy compared with previous methods. Kuang and Wang use deep learning techniques to uncover sequence patterns beyond CTCF motif pairs that mediate the formation of chromatin loops. Their DeepCTCFLoop model is shown to outperform word2vec and boosted trees methods at distinguishing CTCF motif pairs that form chromatin loops from the pairs that do not. Yin et al. develop deep learning methods for diagnosis of autism spectrum disorder from brain functional magnetic resonance imaging data. Experimental results on the Autism Brain Imaging Data Exchange 1 data set show that the proposed deep learning methods outperform previous methods. Syed et al. use Support Vector Machine and Random Forest classifiers to predict adherence to the treatment guidelines from the National Comprehensive Cancer Network based on both clinical and nonclinical factors. The authors find that the center in which the patient was treated played a significant role in adherence to the guidelines and whether or not androgen deprivation therapy was prescribed.
The remaining articles cover diverse applications in computational biology, from structural variation identification and data integration to modeling of single molecule sequencing fluorescence signals and firing rates of neurons. Hayes et al. present a new algorithm based on clique partitioning for identifying complex genomic structural variants from whole genome sequencing data. Experiments on both simulated and real data sets demonstrate the effectiveness of their algorithm, called CleanBreak. Thapa and Ali describe a graph database model that can be used to store multiomics data including gene and miRNA expression, DNA methylation, point mutations, copy number variations, and tissue slide images, and demonstrate its use on data sets from the Cancer Genome Atlas. Chen and Lu present statistical models and decoding algorithms for fluorescence signals captured in single molecule sequencing. Among others, their model can yield improved sequencing accuracy by optimizing fluorescence dye selection. Finally, the article by Cheng and Lu introduces the concept of neuron agility, which describes how fast a neuron can respond to a periodic input signal. Agility score functions are derived for three neuron models enabling the determination of the angle of phase shift for a given input signal frequency.
We thank the Co-Editors-in-Chief, Sorin Istrail and Michael S. Waterman, for providing us with the opportunity to showcase some of the exciting research presented at ICCABS 2019 in the Journal of Computational Biology. Last but not least, we thank all ICCABS authors—the conference could not continue to thrive without their high-quality contributions.