Modeling and Simulation of Biological Regulatory Networks using FPGA

Abstract

The study of biological regulatory networks constitutes an important area in systems biology. Several modeling approaches have been introduced to describe the behavior of these networks. Qualitative modeling approach presented by Ren´e Thomas has been widely used to discover dynamic properties of biological networks. A number of software tools have been developed to study and analyze biological networks using this formalism . However due to large parameter space, huge amount of computational resources are required to study large networks. In last few years, reconfigurable hardware devices such as field programmable gate arrays (FPGA) have demonstrated huge computational power. Due to their high compu tational capability and low power consumption, FPGA offer attractive platform to develop computational applications. In this work, we implement Ren´e Thomas modeling approach on FPGA. We used verilog language for FPGA programming. A C++ application programming interface (API) is developed that reads input file to extract biological network and generate verilog code on runtime time for FPGA. The important qualitative properties of biological networks such as stable states and oscillations (cycles) are computed on FPGA. The experiments were simulated for Virtex-6 FPGA ML605 using Xilinx ISE Design Suite version 13.1. This design checks sysntax of program,simulate verilog code and generate top level module and circuit diagrams. The experimental results, when compared to desktop computers, show huge performance gain and dramatic reduction in processing time.