Biology Reference
In-Depth Information
Contact distribution ROC
Δ
H profile and contact map
1.0
Δ
0.0: 0.706, 0.792 (19)
Δ H 0.0: -0.139, 0.208 (5)
H
0.005
Δ
(109)
Ligand (24)
Ion (5)
H
0.004
0.003
0.002
0.001
0.8
0.6
0.4
00
-0.001
-0.002
-0.003
0.2
0.0
Residue index
0.0
0.2 0.4 0.6
False Positive Rate
contact distribution ROC
0.8
1.0
0
50
100 150
Δ H profile and contact map
200
250
300
1.0
Δ H 0.0: 0.412, 0.895 (17)
0.008
0.006
0.004
0.002
Δ
(109)
Ligand (19)
H
Δ H 0.0: -0.652, 0.105 (2)
0.8
0.6
0.4
00
-0.002
-0.004
-0.006
0.2
Residue index
0.0
0.0
0.2
0.4 0.6
False Positive Rate
0.8
1.0
0
20
40
60
80
100
Fig. 4.17 Examples of proteins for which the FOD model correctly identified binding sites. The
top row presents 2Z6I - its 3D structure (with a clearly visible deep binding pocket) as well as the
ROC curve indicating the relation between TPR and FPR. The area marked in red represents posi-
t i v e
Δ H results while the area marked in blue indicates TPR and FPR coefficients for negative
Δ H values. The attached plots represent
Δ H pro fi les for both proteins ( top - 2Z6I; bottom -
1N9L), with highlighted ligand binding sites, as well as ion binding sites in the case of 1N9L
(which, however, is out of scope of this discussion). 3D images have been derived from PDBSum
(for 2Z6I) and PDB (1N9L)
Δ H profile and contact map
Δ H profile and contact map
Δ
H (174)
Ligand (13)
0.006
0.004
0.002
0.0
-0.002
Δ
H (339)
Ligand (24)
0.004
0.002
0.0
-0.004
-0.002
-0.006
-0.008
-0.004
Residue index
0
20
40
60
80
100
120
140
160
0
50
100
150
200
250
300
Residue index
Δ H
profile and contact map
0.004
0.003
0.002
0.001
0.0
Δ
H (327)
Ligand (17)
Δ
H (446)
Ligand (18)
0.004
0.002
0.0
-0.001
-0.002
-0.002
-0.003
-0.004
-0.004
0
50
100
150
200
250
300
350
400
0
50
100
150
200
250
300
Residue index
Residue index
Δ H pro fi les, with ligand binding residues marked by green triangles and catalytic residues
marked by red circles
Fig. 4.18
Search WWH ::




Custom Search