Design and Analysis of Rotator Core for Short Reads Sequence Alignment Using Burrows-Wheeler Algorithm

This paper presents the design and analysis of rotator core for short reads sequence alignment using Burrows-Wheeler algorithm. This rotator core aims to reduce alignment time and improve performance, while at the same time achieving a high level of configurability and accuracy. The alignment of millions of short DNA fragments to a large genome is an aspect that is highly crucial to the world these days, especially contributing towards modern computational biology. Improvements in DNA sequencing alignment would increase the size of sequencing datasets while also broadening their utility. Unfortunately, the processing of DNA sequence alignment takes quite some time, with many hours to complete. This paper aims to model both hardware and software of the rotator core using Verilog. Afterwards, designs will be synthesized and simulated using Quartus, Design Compiler and Cloud V. Based on results and discussions, all developed rotator designs were successfully synthesized in Quartus. RTL netlist and waveform in Cloud V for rotator core designs were successfully generated and simulated. In Design Compiler, the normal compile option proved to be better than the compile power and compile_ultra options with respect to timing max delay. However, the compile_power option was found to be better than the normal compile and compile_ultra options in terms of timing min delay. For area, the compile_power option was also found to be better than other options, while the compile_ultra option becames the better option in terms of power. In conclusion, the design of rotator cores for short reads sequence alignment was proven to be successful.