Notesale: Turn your study into money

Document Preview

Extracts from the notes are below, to see the PDF you'll receive please use the links above

---

BIOC495 - Pairwise Sequence Alignment
Why can aligning a protein sequence be more informative than DNA sequence? =1
...
AAs share related biophysical properties
3
...
Confirming identity of DNA
sequence in database search
2
...
Analyzing identity of cloned cDNA fragment
4
...
score for creating gap
2
...
A gene undergoes DNA
mutation that it encodes a different AA
2
...
likely
Which PAM matrix is appropriate for comparisons of closely related proteins? =PAM10
Which matrix yields the highest score for relatively divergent proteins? =PAM70
Why do lower PAM matrices (e
...
PAM10 and PAM60) produce lower bit scores? =Low AA identity
Mismatches assigned large negative scores
What does BLOSUM stand for? =blocks substitution matrix
What is the BLOCKS database? =consists of over 500 gorups of local multiple alignments (blocks) of
distantly related protein sequences
The BLOSUM scoring scheme employs what? =log-odds ratio using base 2 logarithm
What is the default scoring matrix for BLAST protein search programs at NCBI? =BLOSUM62
What does BLOSUM62 do? =Merges all proteins in an alignment that has 62% amino acid identity or
greater into one sequence

SO: It is useful to use when comparing sequences with LESS than 62% identity
Are substitution frequencies for the BLOSUM62 matrix are weighted (more/less) heavily by blocks of
protein sequences having less than 62% identity =MORE
What are the other two BLOSUM matrices commonly used? =BLOSUM50 and BLOSUM90

So if aligning two sequences with around 50% identity, use BLOSUM50
What is the similarity between BLOSUm and PAM matrices? =both use log=odds values in scoring
systems
What data are PAM matrices based on? =data from alignment of CLOSELY RELATED protein families

Assumption that substitution probabilities for highly related proteins can be extrapolated to probabilities
for distantly related proteins
What are BLOSUM matrices based on? =empirical observations of more distantly related protein
alignments
When may the PAM30 matrix be useful? =for identifying significant conservation between two closely
related proteins
When is a BLOSUM80 matrix not suitable? =When scoring closely related sequences

B/c adapted to regions of sequences that share UP TO 80% identity
Beyond that limited region two proteins may share dramatically less AA identity
What are high-value BLOSUM matrices and low-value PAM matrices best suited for? =Well conserved
proteins
What are low value BLOSUM matrices and high PAM numbers best suited for? =detect distantly related
proteins
What is the best way to determine the detection limits of pairwise alignments? =statistical tests that
assess the likelihood of finding a match by chance
Is percent identity an exact indicator of the number of mutations that have occurred across a protein
sequence? =nah
What does the PAM250 matrix assume? =Assumes the occurrence of 250 point mutations per 100 amino
acids
What is the "twilight zone" of pairwise alignment? =AT 250 point mutations per 100 AAs --> difficult to
assess if two proteins are homologous
What is a global alignment? =Contains entire sequence of protein or DNA molecule
Uses method of Needleman and Wunsch
What is local alignment? =Focuses on regions of greatest similarity between two sequences
Method of Smith and Waterman

What is one approach to deciding whether two sequences are significantly related from an evolutionary
POV? =consider percent identity
Helps obtain sense of degree or relatedness
Is percent identity alone sufficient to demonstrate/rule out homology? =NO -High degree of identity over short region might not be evolutionarily significant
Low percent identity could reflect homology
What is the general cutoff for percent identity to demonstrate they are related? =25% or more AA
identity over span of 150 or more AAs
What percent identity does 70 AA residues resonable to estimate the proteins are homologous? =40%



---