Scribe: D. Slotta

Today's Handouts and Announcements

• None

Today's Topics

• Biologist <-> Computer Scientist Interaction
  • Old Method: Biologist - problems -> Computer Scientist - models, algorithms -> publication

  • New Method: Biologist - problems, data, guidance -> Computer Scientist - results -> Biologist

• Algorithms: Solve problems computationally
  • Bioinformatics (usually) deals with discrete problems.
  • Problems have instances (input) and solutions (output).
  • Representations for instances and solutions are strings over a finite alphabet.
  • An algorithm solves a problem P if it is an implementation of P under some model of computation.

• Problems
  1. Pattern matching (decision problem)
  2. Travel routing (neither)
  3. Sorting (neither)
  4. Shortest Path (optimization problem)
  5. Traveling salesman (optimization problem)

• Model of Computation
  • Characteristics
    1. Get input
    2. Generate output
    3. Has state
    4. State changes occur in steps

  • Examples
    1. Turing machine
    2. Circuits of logic gates
    3. Random access machine (RAM)
    4. Parallel model (PRAM)

• Pseudocode: representation for algorithms

  Statement Type	Example
  Read Input	read i, j, k
  Generate Output	write j, k
  Assignment Statement	x <- 17
  Conditional	if ( x=17 ) then ... else ...
  Iteration	for, while, repeat
  Recursion	name piece of pseudocode, give parameters, call it
  Data Structures	Everything is an integer, or array of integers

Today's Sources

• Setubal and Meidanis: Chapter 2.

• Salzberg, Searls, and Kasif: Chapter 2.