Login

Join for Free!
119273 members
table of contents table of contents

A simple physical energy function, which uses electrostatics, solvation, hydrogen bonds and …


Biology Articles » Biophysics » Molecular Biophysics » Protein–DNA binding specificity predictions with structural models » Tables

Tables
- Protein–DNA binding specificity predictions with structural models

....................................................................................................

Table 1 Experimental binding affinity dataset

Name PDB code Method (Res., Å) {Delta}{Delta}G data points Organism Reference
Zif268 1aay X-ray (1.6) 15 (8) Mus musculus (42)
Zif268 1aay X-ray (1.6) 6 (6) M.musculus (67)
Zif268 D20A 1jk1 X-ray (1.9) 6 (6) M.musculus (67)
Tus 1ecr X-ray (2.7) 20 (20) Escherichia coli (68)
LacR 1efa X-ray (2.6) 5 (5) E.coli (69)
{lambda}R 1lmb X-ray (1.8) 51 (51) {lambda}-Phage (51)
TrpR 1tro X-ray (1.9) 9 (9) E.coli (70)
ER 1hcq X-ray (2.4) 7 (7) Homo sapiens (71)
CroR 6cro X-ray (3.0) 56 (56) {lambda}-Phage (52)
EcoRI 1ckq X-ray (1.85) 13 (13) E.coli (41)
Crp 1run X-ray (2.7) 15 (15) E.coli (72)
BamHI 1bhm X-ray (2.2) 23 (0) Bacillus amyloliquefaciens (49)
PU.1 ETS 1pue X-ray (2.1) 25 (0) M.musculus (50)
Ndt80 1mnn X-ray (1.4) 26 (0) Saccharomyces cerevisiae (34)
MAT a1/{alpha}2 1yrn X-ray (2.5) 54 (0) S.cerevisiae (30)
c-Myb 1mse NMR (NA) 27 (0) M.musculus (47)
AtERF1 1gcc NMR (NA) 21 (0) Arabidopsis thaliana (45)

For protein–DNA structures solved by X-ray crystallography, resolution (Å) is shown in parentheses. The total number of {Delta}{Delta}G measurements are shown for each dataset, with the number of data points used in free energy function parameterization listed in parentheses.

....................................................................................................

Table 2 Experimental binding site and weight matrix dataset

Name PDB code Method (Res., Å) Nseq Organism Reference
{lambda}R 1lmb X-ray (1.8) a {lambda}-Phage (51)
CroR 6cro X-ray (3.0) a {lambda}-Phage (52)
AtERF1 1gcc NMR (NA) a A.thaliana (45)
c-Myb 1mse NMR (NA) a M.musculus (47)
Zif268 1aay X-ray (1.6) 6b M.musculus (31)
Ndt80 1mnn X-ray (1.4) 8b S.cerevisiae (34)
Gcn4p 1ysa X-ray (2.9) 9c S.cerevisiae (35,36)
MAT a1/{alpha}2 1yrn X-ray (2.5) 19c S.cerevisiae (30)
EcR/Usp 1r0o X-ray (2.24) 33c Drosophila melanogaster (57)
Ttk 2drp X-ray (2.8) 16c D.melanogaster
Prd(homeo) 1fjl X-ray (2.0) 15c D.melanogaster (37)
Ubx/Exd 1b8i X-ray (2.4) 4b D.melanogaster
Trl 1yui NMR (NA) 5c D.melanogaster
MetJ 1mj2 X-ray (2.4) 16c E.coli (32)
TrpR 1tro X-ray (1.9) 15c E.coli (32,33)
PhoB 1gxp X-ray (2.5) 16c E.coli (32)
Ihf 1ihf X-ray (2.5) 27c E.coli (32)
DnaA 1j1v X-ray (2.1) 9c E.coli (32)
PurR 2puc X-ray (2.7) 23c E.coli (32)
Crp 1run X-ray (2.7) 50c E.coli (32)

For protein–DNA structures solved by X-ray crystallography, resolution (Å) is shown in parentheses. Nseq is the total number of aligned binding site sequences (including the DNA sequence from the protein–DNA complex).

aPWM is obtained from {Delta}{Delta}G data for all one-point mutations of the binding site.

bPWM is obtained separately from the binding sites listed in the table, by SELEX experiments or independently published surveys of genomic sites.

cAlignment of binding sites listed in the Nseq column is used to create an experimental PWM (binding site from the protein–DNA structure is not included into PWM; pseudocounts are set to 0.0).

....................................................................................................

Table 3 {Delta}{Delta}G predictions summary

N r F {varepsilon} Description
1 0.57 143/196 1.58 static model: full
2 0.57 125/196 1.80 static model: protein–DNA only
3 0.43 133/196 2.90 dynamic model: full
4 0.41 128/196 2.93 dynamic model: protein–DNA only
5 0.39 117/196 1.86 dynamic model: full, fitted weights
6 0.42 137/196 1.73 contact model

r, Linear correlation coefficient; F, fraction of successful predictions (see text); and {varepsilon} (kcal/mol), average unsigned error between predicted and experimental-binding affinities.

....................................................................................................

Table 4 Summary of binding site energy predictions

PDB ZPDB <Zsite> Rank L
1mnn –3.69 –2.54 14 12
1ysa –3.38 –2.53 1 7
1yrn –4.30 –3.34 19
1aay –3.70 –3.60 13 10
1b8i –3.22 –2.74 8 10
1r0o –3.80 –2.88 15
2drp –3.77 –2.76 6 11
1yui –2.38 –2.48 91 7
1fjl –3.22 –2.93 4507 13
1gxp –4.17 –3.06 20
1ihf –3.41 –1.12 34
1j1v –4.20 –3.36 13 13
1mj2 –3.09 –2.13 16
1run –3.50 –2.32 22
1tro –3.73 –2.57 18
2puc –3.88 –3.68 16

ZPDB is the Z-score (Equation 16) for the protein–DNA binding energy with the binding site found in the protein–DNA structure; <Zsite> is the average Z-score for protein–DNA binding energies with binding sites listed in Table 2; Rank is the rank of the binding energy for the structural site in the ensemble of 4L sequences (L is the binding site length). Rank was computed for all binding sites with L

....................................................................................................

Table 5 The {psi}-test of PWM predictions for the contact, static, dynamic and random models

PDB {psi}contact {psi}static {psi}dynamic <{psi}random>
1mnn 0.14 0.12 0.20 0.68
1ysa 0.16 0.31 0.39 0.91
1yrn 0.20 0.26 0.36 0.73
1lmb 0.10 0.09 0.14 0.47
6cro 0.07 0.10 0.21 0.47
1mse 0.26 0.24 0.55
1gcc 0.15 0.12 0.57
1aay 0.13 0.19 0.35 0.95
1b8i 0.35 0.34 0.36 0.87
1r0o 0.22 0.25 0.38 0.72
2drp 0.19 0.24 0.23 0.69
1yui 0.07 0.26 0.95
1fjl 0.30 0.32 0.51 0.83
1gxp 0.24 0.28 0.41 0.68
1ihf 0.21 0.19 0.43
1j1v 0.21 0.22 0.36 0.70
1mj2 0.21 0.38 0.33 0.69
1run 0.10 0.17 0.38 0.51
1tro 0.30 0.31 0.39 0.71
2puc 0.18 0.26 0.63 0.81

Random model uses randomly sampled normalized PWM entries (see Methods). Three NMR structures and Ihf (1ihf) are excluded from the dynamic model predictions.

 ....................................................................................................

 


rating: 3.00 from 2 votes | updated on: 16 Dec 2007 | views: 8643 |

Rate article:







excellent!bad…