Improving the efficiency of a user-driven learning system with reconfigurable hardware. Application to DNA splicing

Area of Science:

• Computational Biology
• Computer Engineering
• Bioinformatics

Background:

• Existing methods for genetic database scanning lack efficiency.
• Human-expert and machine interaction models have been explored separately.
• Reconfigurable hardware systems offer high-performance computing capabilities.

Purpose of the Study:

• To develop a hybrid approach for problem-solving by integrating software and hardware components.
• To investigate the efficiency of human-expert and machine interaction within a reconfigurable hardware framework.
• To apply the developed system to a specific bioinformatics challenge: primate splice junction recognition.

Main Methods:

• Combining two prior works: a conceptual environment for human-machine interaction and a high-performance genetic database scanning algorithm using reconfigurable hardware.
• Implementing a system where concept modeling is based on a reconfigurable hardware system.
• Facilitating real-time interaction speeds between human experts and the machine.

Main Results:

• Demonstrated efficient human-expert and machine interaction through concept modeling on reconfigurable hardware.
• Achieved real-time interaction speeds for cooperative problem-solving.
• Partially applied the system to recognize primate splice junction sites in genetic sequences, showing successful partial application.

Conclusions:

• The integration of human expertise with reconfigurable hardware-based concept modeling enables efficient, real-time problem-solving.
• This hybrid approach shows promise for complex bioinformatics tasks like genetic sequence analysis.
• The system's successful partial application to primate splice junction recognition validates its potential.