Skip to content

Latest commit

 

History

History
15 lines (10 loc) · 849 Bytes

testingDiscussion.md

File metadata and controls

15 lines (10 loc) · 849 Bytes

Thoughts about our unit tests

Things that our random dna generator might do poorly but we don't know how to test for:

  • Might not really be random...:
    • Might spit out all 'C's repeatedly, every time. WLOG T or G or A.
    • Might spit out 'ACTG' over and over again, every time. Or some other pattern.
  • Might not have the patterns or distributions of real DNA. Aren't there stats packages we could use for this?

Things that our random dna slicer might do poorly but we don't know how to test for:

  • Might not slice the way a real life slicer does:
    • Might always slice a sequence at the same places so we get no overlaps. This makes re-sequencing impossible.

Our simulation of mis-reading ("corruptTheSequences( )" and "corruptTheSequence( )") might not get the kinds of mistakes in reading that real sequencing equipment makes.