講習会スライド.key

Size: px

Start display at page:

Download "講習会スライド.key"

おさむさかいざわ
5 years ago
Views:

1 ef-site/ef-surf/ef-seek PDBj2009

2 Today s Topic++ ef-site ef-surf ef-seek 2

@2009/04/19 @2009/07/20:3187 http://www.

3 3

4 ATGC MKRLD 4

5 3today s topic or 5

6 : 6

7 7

8 (1.4Å) 8

9 Poisson Poisson-Boltzmann εp = 2.0 εs = 80.0 Laplace

10 PDB NMR ef-site ID 10

11 Top page OR PDBj 11

12 jv DB 12

13 jv formerly known as PDBjViewer Java + JOGL RasmolRasmol-like stand aloneapplet PDB-ML, PDB, polygon ( ) xpsss windows XP (sp2), Mac OS-X (10.3), vine3.1fedora core 3Linux OpenGL supported graphic card and latest graphic driver 13

14 14

15 Myb proto-oncogene protein DNA-bind DNA-bind DNA-bind DNA-bind 15

16 seqinfo XML efvet-ml XML jv molscript file ( efvet-flat XML 16

17 Molscript file Molscript (Kraulis, J. Appl. Cryst, v24, p , 1991) PDB Object ex) molscript input file 1rop.objobject file 17

18 1) PDBupload 2) 3URL 4) 15min-60min 18

19 ef-seekhttp://ef-site.hgc.jp/ef-seek ef-site PDB 19

Normalization of similarity score N hetero compound binding sites appeared

Z-score = (score mean)/std coverage More significant z-score similarity

Normalization from functional sites view Query protein coverage Results

20 Normalization of similarity score N hetero compound binding sites appeared in PDB Functional site patches Normalization from query protein s view. Z-score = (score mean)/std coverage More significant z-score similarity score = Larger patches would get larger Z-score. Normalization from functional sites view Query protein coverage Results will be shown in coverage vs. Z-score plot. The number of corresponding vertexes is used as similarity score. 20

21 Threshold line determination 10 randomly selected representative with free and complex structures. Homologous proteins with similar ligands are considered to be correct. Ethylene glycol Glycerol 2,5-dimethyl- pyrimidin-4- ylamine Myo-inositol Castanospermine N-acetyl-dgalactosamine (Hydroxy ethyloxy)tri (ethyloxy)octane Praziquantel ADP NAD 21

22 Threshold line determination Maximize CC with a threshold line cov. = under a constraint that fraction of TP exceed 90%, 70% or 50% 90%-TP line will be used hereafter. CC = (TP TN FP FN) (TP + FP)(TP + FN)(TN + FP)(TN + FN) 22

23 Modestly promising Very highly promising 90%-TP 50%-TP Maybe promising View complex 23

24 24

DNA Methionine Repressor (1MJQ) Predicted site

positive Yellow: hydrophobic surface Local curvature

Binding site Red: protrusion, Blue: concave Global

25 DNA Methionine Repressor (1MJQ) Predicted site Electrostatic potential Red: negative, blue: positive Yellow: hydrophobic surface Local curvature relating to DNA-binding directly: slightly protruded Binding site Red: protrusion, Blue: concave Global curvature relating to entire surface geometry: largely concave 3 25

26 protrusion concave Saddle points Mean Curvature Gaussian Curvature = (k max +k min )/2 = k max k min

27 Mean Curvature(H) Gaussian Curvature(K) = (k max +k min )/2 = k max k min 27

Prediction Scheme-1: Statistical Preference Measure Relative Frequency F bind (φ e, K local, K global ) = N bind (φ e, K local, K global ) / N bindtotal F non-bind (φ e, K local, K global ) = N

28 Prediction Scheme-1: Statistical Preference Measure Relative Frequency F bind (φ e, K local, K global ) = N bind (φ e, K local, K global ) / N bindtotal F non-bind (φ e, K local, K global ) = N non-bind (φ e, K local, K global ) / N non-bindtotal Distribution of electrostatic potential, local curvature and global curvature for all proteins in dataset-1 Statistical Preference Measure P bind / P non-bind 28

29 Prediction Scheme-2: Prediction Score (Pscore) Statistical Preference Measure P bind / P non-bind > 4.0 For each vertex, calculate the measure and colour it when the value exceed 4.0. Pscore = max (Parea / Whole area) P bind / P non-bind Parea : predicted DNA-binding weighted area for a given direction Whole area: whole weighted area for a given direction. Weights are calculated as inner product of normal vector and direction vector. Maximization was done by searching all possible direction by 10 interval. Direction vector Pscore will be used as an indicator of the prediction results. 29

30 Prediction Results Tsuchiya et al. (2004) PROTEINS, 55, relative frequency P score Histogram of Prediction score for dsdna-binding proteins (63), ATPbinding proteins (21), and non-dsdna-binding proteins (406) 86% accuracy for predicting dsdna-binding proteins, and 96% accuracy for predicting non-dna-binding proteins including ATP-binding proteins. 30

31 PreDs: Predition of DNA-binding site 31

32 Disorder region Disorder (Dyson, HJ and Wright, PE, 2005) 32

33 disorder Dunker et.al, Biochemistry, (Molecular recognition) DNA 3. -RNA (t, r, m) (Molecular assembly) 3. (Protein modification) 90 (Dunker et.al, Biochemistry, 2002) DisProt ( 469 proteins, 1114 regions 33

34 Disordered Kingdom # of proteins Disorder freq. (% of aa) Length > 30 (% of chains) Length > 50 (% of chains) Archaea 11, Bacteria 35, Eukaryota 88, Ward et al, JMB, 337, , 2004 Estimation by DISOPRED2 (Jones et al) 34

35 Ishida & Kinoshita, 2007, 2008 Support Vector Machine (PSSM) Meta 10-fold cross validation meta-predictor ROC curve 35

36 Complex form (1mqw) Free form (1mp2) Green parts is flexible and invisible in the free form of the structure, which could prevent our method to predict the binding site. 36

37 PDB Missing loops (or gaps) are identified by comparing SEQRES and ATOM record in each PDB file. N-terminal and C-terminal gaps are ignored. 7,949 loops are invisible among 41,417 chains in current PDB Apr, 2005 Entries with 2.5 or better resolution About 63% missing loop are 8 or less residue long 8 residues are said to be the threshold value to build the good model. Length distribution of missing loop in PDB 37

38 IDP: PrDOS Glu... etc (herg Predict 38

39 ef-site ID PID_ModelID-ChainID : 1a1t_3-A ModelID : 1tup-C Chain : 1tup-ABCEF ef-site ID PDB-ID 39

40 seqinfo 40

41 efvet-ml 41

講習会スライド.key

講習会スライド.key PDBj 2012 2 7 kengo@ecei.tohoku.ac.jp http://www.sb.ecei.tohoku.ac.jp ef-site ef-surf ef-seek 2 2011/01/04 2011/12/20: 12 765 http://www.ncbi.nlm.nih.gov/genomes/static/gpstat.html 3 ATGC MKRLD 4 3 today