No abstract

BackgroundHuman Down syndrome (DS) is usually caused by genomic micro-duplications and dosage imbalances of human chromosome 21. It is associated with many genomic and phenotype abnormalities. Even though human DS occurs about 1 per 1,000 births worldwide, which is a very high rate, researchers haven’t found any effective method to cure DS. Currently, the most efficient ways of human DS prevention are screening and early detection.MethodsIn this study, we used deep learning techniques and analyzed a set of Illumina genotyping array data. We built a bi-stream convolutional neural networks model to screen/predict the occurrence of DS. Firstly, we built image input data by converting the intensities of each SNP site into chromosome SNP maps. Next, we proposed a bi-stream convolutional neural network (CNN) architecture with nine layers and two branch models. We further merged two CNN branch models into one model in the fourth convolutional layer, and output the prediction in the last layer.ResultsOur bi-stream CNN model achieved 99.3% average accuracies, and very low false-positive and false-negative rates, which was necessary for further applications in disease prediction and medical practice. We further visualized the feature maps and learned filters from intermediate convolutional layers, which showed the genomic patterns and correlated SNPs variations in human DS genomes. We also compared our methods with other CNN and traditional machine learning models. We further analyzed and discussed the characteristics and strengths of our bi-stream CNN model.ConclusionsOur bi-stream model used two branch CNN models to learn the local genome features and regional patterns among adjacent genes and SNP sites from two chromosomes simultaneously. It achieved the best performance in all evaluating metrics when compared with two single-stream CNN models and three traditional machine-learning algorithms. The visualized feature maps also provided opportunities to study the genomic markers and pathway components associated with Human DS, which provided insights for gene therapy and genomic medicine developments.

No abstract

No abstract

BackgoundEvolution of cancer cells is characterized by large scale and rapid changes in the chromosomal  landscape. The fluorescence in situ hybridization (FISH) technique provides a way to measure the copy numbers of preselected genes in a group of cells and has been found to be a reliable source of data to model the evolution of tumor cells. Chowdhury et al. (Bioinformatics 29(13):189–98, 23; PLoS Comput Biol 10(7):1003740, 24) recently develop a computational model for tumor progression driven by gains and losses in cell count patterns obtained by FISH probes. Their model aims to find the rectilinear Steiner minimum tree (RSMT) (Chowdhury et al. in Bioinformatics 29(13):189–98, 23) and the duplication Steiner minimum tree (DSMT) (Chowdhury et al. in PLoS Comput Biol 10(7):1003740, 24) that describe the progression of FISH cell count patterns over its branches in a parsimonious manner. Both the RSMT and DSMT problems are NP-hard and heuristics are required to solve the problems efficiently.MethodsIn this paper we propose two approaches to solve the RSMT problem, one inspired by iterative methods to address the “small phylogeny” problem (Sankoff et al. in J Mol Evol 7(2):133–49, 27; Blanchette et al. in Genome Inform 8:25–34, 28), and the other based on maximum parsimony phylogeny inference. We further show how to extend these heuristics to obtain solutions to the DSMT problem, that models large scale duplication events.ResultsExperimental results from both simulated and real tumor data show that our methods outperform previous heuristics (Chowdhury et al. in Bioinformatics 29(13):189–98, 23; Chowdhury et al. in PLoS Comput Biol 10(7):1003740, 24) in obtaining solutions to both RSMT and DSMT problems.ConclusionThe methods introduced here are able to provide more parsimony phylogenies compared to earlier ones which are consider better choices.