4. Databases and comparisons

Question 1

Databases with protein data

Answer 1

UniProt
InterPro
Pride
PDB

Question 2

Databases for genome and genes

Answer 2

Ensembl

Question 3

Databases of sequence info

Answer 3

Uniprot/swissprot
uniprot/trembl
EMBL (nucleotide sequence)
Genbank

Question 4

Swissprot

Answer 4

• high level of annotation eg function, domains, PST
• minimal level of redundancy
• quality of annotation

Question 5

EMBL (embl-ebi)

Answer 5

• have eg emboss needle, interpro etc links
• sequences submitted directly by scientists
• literature and patterns
• little error checking

Question 6

Ensembl

Answer 6

• useful when analysing genomes

have gene and find mammalian homologes or eg all known SNPs

Question 7

why sequence comparison

Answer 7

• identification of protein
• search for homologies
• evolution

Question 8

Methods for sequence comparisons

Answer 8

• Diagonal plot
• BLAST
• FASTA

Question 9

FASTA

Answer 9

• use scoring matrices
• use k-tuples
  
  • look at more than one residue at the time
• hashing
  
  • make dictionary/table of k-tuples
• find clusters of k-tuples
  
  • then you do dynamic programming

Question 10

BLAST

Answer 10

• faster than FASTA, also more sensitive now a days
• use “words” instead of k-tuples
  
  • instead of identical hits you use score threshold
• generally it looks at longer sequences

Question 11

E-value

Answer 11

a parameter that describes the number of hits one can “expect” to see by chance when searching a database of a particular size. Essentially, the E value describes the random background noise. The lower the E-value, or the closer it is to zero, the more ”significant” the match is.