Comparison of Multiple Protein Sequence Alignment Algorithms


Abstraction The idea behind my project is to compare multiple protein sequence alignment algorithms on speed and accuracy. The algorithms being compared will encompass a wide range of methods for finding the alignment, however I will personally be implementing one of these algorithms which relies on segment to segment comparison instead of the normal residual to residual comparison. Also I will be developing one or two modified versions of this algorithm to try to increase the speed or accuracy even at the expense of the other. What I am aiming for is to determine how well the algorithm being defined within the cited paper can be constructed into a functioning program along with the efficiency this algorithm present when compared to other published methods. However, the a side objective is to see if I am able to implement changes in the algorithm that will allow the algorithm to improve to the point that it overtake one of the algorithms being compared or simple outperforms the current algorithm.
Team members Justin Currey
Itemized
Plan of Action
  • Read and fully understand "Multiple DNA and protein sequence alignment based on segment-to-segment comparison".
  • Use Protein Data Bank to generate random sets of proteins to compare for beta testing.
  • Implement and test algorithm in cited paper.
  • Research different algorithms to determine valid changes I can make to improve the algorithm.
  • Implement and test 1-2 new algorithm variations aimed towards speed, accuracy, or both.
  • Find 2-4 working algorithms for multiple segment alignment *Exact algorithms to be inserted later.
  • Use Protein Data Bank to generate 20,000 random sets of proteins to use for data comparison.
  • Measure the average speed, acurracy, and overall performance for each algorithm.
  • Compare outputs for all algorithms running a variety of set sizes and sequence similarities.



    		•
    Works Cited Morgenstern B., Dress A., and Werner T. "Multiple DNA and protein sequence alignment based on segment-to-segment comparison" Proc. Natl. Acad. Sci. USA 1996; 93: 12098-12103