Design Pattern Matching Techniques for Next-Generation RNA Sequencing Applications

In recent years, more and more next-generation sequencing techniques have been proposed, such as Roche 454, Illumina Solexa, and ABI SOLiD. The feature of these sequencing techniques is to produce a large amount of short reads for DNA, RNA and small non-coding RNA sequencing. By these reads, we can do the tasks of sequence assembly, gene prediction, cancer analysis and etc. Due to the short length and large number of reads, the previous tools may be not suitable for these reads. The goal of this integrated project is to develop a high throughput sequencing platform to provide techniques and tools from sequencing, functional analysis to gene network analysis. The goal of this sub-project is to design a series of pattern matching techniques for next-generation RNA sequencing, and then applied these techniques to study the gene prediction problems. First year, we will design exact and approximation pattern matching techniques for next-generation RNA sequencing by considering the data preprocessing, speed, and accuracy. We also will modify the proposed techniques to solve the split problem. We will design the pattern matching techniques based on the single-GPU environment. Second year, we will modify the proposed techniques by considering the annotations in EST database. We will apply the proposed techniques to study the gene prediction problem in a single-species. We will design the pattern matching techniques based on the multi-GPUs environment. Third year, we will modify the proposed techniques by considering the problems in the comparative genomics. We will apply the proposed techniques to study the gene prediction problem in multiple-species. We will design the pattern matching techniques based on the GPU-cluster environment.

Project ID:PB10101-3622
External Project ID:NSC100-2221-E182-057-MY3