Review 1

Blast

  • Table:
    • Split into kmers as keys
    • Entries into hash table are storing where the kmers are in the original sequence
ATT(s1,1), (s3,1)
TTA(s1, 2), (s3, 3)
TAT(s1, 3), (s2, 1), (s3, 4)
ATG(s2, 2)
TGT(s2, 3)
…

Bayes Method

K Means

  • Find smallest distance to each cluster
  • Reassign each cluster to average of things closest to it
  • Algo ends when clusters dont move

Genome Assembly
Suffix Tree
UPGMA

de Bruin graph to find the Hamiltonian path

  • kmers of 3:
  • ACTAG: (ACT) -> (CTA) -> (TAG)
  • Create nodes of length 3 that point to a possible following sequence