Showing papers by "Zhong Wang published in 2016"

Reconstructing single genomes from complex microbial communities

Abstract: Abstract High throughput next generation sequencing technologies have enabled cultivation-independent approaches to study microbial communities in environmental samples. To date much of functional metagenomics has been limited to the gene or pathway level. Recent breakthroughs in metagenome binning have made it feasible to reconstruct high quality, individual microbial genomes from complex communities with thousands of species. In this review we aim to compare several automated metagenome binning software tools for their performance, and provide a practical guide for the metagenomics research community to carry out successful binning analyses.

A Scalable Pipeline for Transcriptome Profiling Tasks with On-Demand Computing Clouds

Abstract: We introduce a pilot-based approach with which scalable data analytics essential for a large RNA-seq data set are efficiently carried out. Major development mechanisms, designed in order to achieve the required scalability, in particular, targeting cloud environments with on-demand computing, are presented. With an example of Amazon EC2, by harnessing distributed and parallel computing implementations, our pipeline is able to allocate optimally computing resources to tasks of a target workflow in an efficient manner. Consequently, decreasing time-to-completion (TTC) or cost, avoiding failures due to a limited resource of a single node, and enabling scalable data analysis with multiple options can be achieved. Our developed pipeline benefits from the underlying pilot system, Radical Pilot, being readily amenable to scalable solutions over distributed heterogeneous computing resources and suitable for advanced workflows of dynamically adaptive executions. In order to provide insights on such features, benchmark experiments, using two real data sets, were carried out. The benchmark experiments focus on the most computationally expensive transcript assembly step. Evaluation and comparison of transcript assembly accuracy using a single de novo assembler or the combination of multiple assemblers are also presented, underscoring its potential as a platform to support multi-assembler multi-parameter methods or ensemble methods which are statistically attractive and easily feasible with our scalable pipeline. The developed pipeline, as manifested by results presented in this work, is built upon effective strategies that address major challenging issues and viable solutions toward an integrative and scalable method for large-scale RNA-seq data analysis, particularly maximizing merits of Infrastructure as a Service (IaaS) clouds