Transcript Molecular Biology Databases

Investigation of factors affecting prediction of
protein-protein interaction networks by phylogenetic
profiling
Anis Karimpour-Fard‡ , Ryan T. Gill† , and Lawrence Hunter‡
‡
University of Colorado School of Medicine
†
Department of Chemical and Biological Engineering, University of Colorado, Boulder
[email protected]
http://www.colorado.edu/che/research/faculty/gill/
http://compbio.uchsc.edu/Hunter
Dec 1, 2007
The problem ……
More than 500 Microbial genomes are fully
sequence and there is high percent of genes
with unknown function.
For example:
E. coli K12 15%
P. aeruginosa 45%
http://www.genomesonline.org/
X
The meaning of protein function
C
B
Z
S
A
P
D
The function of protein A is its
action on Substrate to form a
Product
Biochemical view
Y
A
N
M
The function of A is the context of
its interactions with other proteins
in the cell
Post genomic view
Eisenberg, D. et. al. Nature 2000
Prediction protein function
•
Homology based methods (gives partial understanding about protein
role)
– Simple sequence similarity searches (BLAST)
– Profile searches (PSI-BLAST)
– Databases of conserved domains (Pfam, SMART)
• Prediction from genomic context
•
•
•
•
Phylogenetic profile
Gene cluster
Gene neighbor
Rosetta Stone
• Prediction from high-throughput experimental data
– Microarray gene expression data
– Protein-protein interaction screens
– ...
Phylogenetic Profile
Pellegrini et al. PNAS 96, 4285 (1999)
Marcotte et al. PNAS 97, 12115 (2000)
1- Select sets of genomes as a reference set
Reference selection?
Does the selection of the reference genomes influence the prediction?
if so? How?
2- Create phylogenetic profile matrix for target organism:
•Do one-against-all BLAST search to identify all homologous
target genes in diverse reference organisms.
Reference selection
Measure profile similarities
How E-value threshold effects
the protein-protein
interactions prediction?
Blast E-value threshold (present or absent)
Generate Protein-protein interactions network
3- Measure profile similarities
Protein X: 110001111001001110001111
Protein Y: 111000111100000110001111
19 matching bits out of 24
4- Generate protein-protein interactions
2 nodes are connected if the 2
proteins have similar profile)
Protein
X
5- Create clusters from set of protein-protein interactions
6- Visualize network
Protein Y
Measure profile similarities
Protein
X
2 nodes are connected if the 2
proteins have similar profile)
Protein Y
Protein X:
110001111001001110001111
Protein Y:
111000111100000110001111
•Inverse homology
•Calculate the homology
between two genomes:
• The ratio of number of
homologs of each reference
organism j to the number of
proteins in the target
genome i ( Hi,j) .
•Pij =1/( Hi,j) otherwise Pij =0.
Karimpour-Fard et al. BMC Genomics. 2007;8(1):393
•Pearson correlation coefficient
•Mutual information
MI(X,Y) = H(X) + H(Y) - H(X,Y)
H(Y) = -∑p(i) ln p(i)
p(i), (i= 0, 1) as the fraction of
genomes in which protein Y in the
state i
1 1
H(X, Y)     p(i, j ) ln p(i, j )
i  0 j 0
Comparison of different combinations of reference genomes and E-value thresholds
using COG
Aerobic
All
Low GC
Random
sets
•
c)
Karimpour-Fard et al. BMC Genomics. 2007;8(1):393
PPV =TP/(TP+FP)
– TP = # predicted pair in the same functional category
– FP= # predicted pair that were classified but were not same functional category
Co-evolution can be used to assign function to unstudied
genes
Edge color code:
• E. coli
K12 (green)
•E. coli O157 (blue)
•Shigella flexneri (black)
•S. typhimurium LT2 (purple)
•P. aeruginosa (mustard)
Hypothetical proteins YcgB,YeaH,YeaG are co-conserved across
different species. Comparison of sub-graphs across species (CS-CCC)
suggested that a previously unstudied S. typhimurium gene, ycgB, is
functionally related to yeaH. Experimental data support the hypothesis
that both genes are important for antimicrobial peptide resistance.
Karimpour-Fard et al. Genome Biology 2007 8:R185