Presentation is loading. Please wait.

Presentation is loading. Please wait.

CISC667, F05, Lec20, Liao1 CISC 467/667 Intro to Bioinformatics (Fall 2005) Protein Structure Prediction Protein Secondary Structure.

Similar presentations


Presentation on theme: "CISC667, F05, Lec20, Liao1 CISC 467/667 Intro to Bioinformatics (Fall 2005) Protein Structure Prediction Protein Secondary Structure."— Presentation transcript:

1 CISC667, F05, Lec20, Liao1 CISC 467/667 Intro to Bioinformatics (Fall 2005) Protein Structure Prediction Protein Secondary Structure

2 CISC667, F05, Lec20, Liao2 Protein structure Primary: amino acid sequence of the protein Secondary: characteristic structure units in 3-D. Tertiary: the 3-dimensional fold of a protein subunit Quaternary: the arrange of subunits in oligomers

3 CISC667, F05, Lec20, Liao3 Experimental Methods X-ray crystallography NMR spectroscopy Neutron diffraction Electron microscopy Atomic force microscopy

4 CISC667, F05, Lec20, Liao4 Computational Methods for secondary structures –Artificial neural networks –SVMs –… Computational Methods for 3-D structures –Comparative (find homologous proteins) –Threading –Ab initio (Molecular dynamics)

5 CISC667, F05, Lec20, Liao5

6 6

7 7

8 8

9 9 Helix complete turn every 3.6 AAs Hydrogen bond between (-C=O) of one AA and (-N-H) of its 4 th neighboring AA

10 CISC667, F05, Lec20, Liao10 Hydrogen bond b/w carbonyl oxygen atom on one chain and NH group on the adjacent chain

11 CISC667, F05, Lec20, Liao11 Ramachandran Plot PHI: -57; PSI -47

12 CISC667, F05, Lec20, Liao12 Ramachandran Plot Parallel: PHI: -119; PSI: 113 Anti-parallel: PHI: -139; PSI: 135

13 CISC667, F05, Lec20, Liao13

14 CISC667, F05, Lec20, Liao14

15 CISC667, F05, Lec20, Liao15 Residue conformation preferences Helix: A, E, K, L, M, R Sheet: C, I, F, T, V, W, Y Coil: D, G, N, P, S

16 CISC667, F05, Lec20, Liao16 Artificial neural networks Perceptron o(x 1, …, x n ) = g(∑ j W j x j ) ∑ j W j x j g o x1x1 W1W1 Input links Output Input function output Activation function X 0 = 1 W0W0 x2x2 xnxn W2W2 WnWn......

17 CISC667, F05, Lec20, Liao17 Activation functions +1 +1 Sigmoid(x) = 1/(1+e -x ) Sign(x) = 1 if x ≥ 0 -1 otherwise Step(x) = 1 if x ≥ t 0 otherwise t x xx

18 CISC667, F05, Lec20, Liao18 Artificial Neural Networks

19 CISC667, F05, Lec20, Liao19 2-unit output

20 CISC667, F05, Lec20, Liao20 Learning: to determine weights and thresholds for all nodes (neurons) so that the net can approximate the training data within error range. –Back-propagation algorithm Feedforward from Input to output Calculate and back-propagate the error (which is the difference between the network output and the target output) Adjust weights (by gradient descent) to decrease the error.

21 CISC667, F05, Lec20, Liao21 w1w1 w0w0 E[w] Gradient descent w new = w old - r [∂E/∂w] where r is a positive constant called learning rate, which determines the step size for the weights to be altered in the steepest descent direction along the error surface.

22 CISC667, F05, Lec20, Liao22 Data representation

23 CISC667, F05, Lec20, Liao23 Issues with ANNs –Network architecture FeedForward (fully connected vs sparsely connected) Recurrent Number of hidden layers, number of hidden units within a layer –Network parameters Learning rate Momentum term –Input/output encoding One of the most significant factors for good performance Extract maximal info Similar instances are encoded to “closer” vectors

24 CISC667, F05, Lec20, Liao24 An on-line service

25 CISC667, F05, Lec20, Liao25 Performance –ceiling at about 65% for direct encoding Local encoding schemes present limited correlation information between residues Little or no improvement using multiple hidden layers. –Surpassing 70% by Including evolutionary information (contained in multiple alignment) Using cascaded neural networks Incorporating global information (e.g., position specific conservation weights)

26 CISC667, F05, Lec20, Liao26 Cathy Wu, Computers Chem. 21(1997)237-256

27 CISC667, F05, Lec20, Liao27 Resources Protein Structure Classification –CATH: http://www.biochem.ucl.ac.uk/bsm/cath/ –SCOP: http://scop.mrc-lmb.cam.ac.uk/scop/http://scop.mrc-lmb.cam.ac.uk/scop/ –FSSP: PDB: http://www.rcsb.org/pdb/


Download ppt "CISC667, F05, Lec20, Liao1 CISC 467/667 Intro to Bioinformatics (Fall 2005) Protein Structure Prediction Protein Secondary Structure."

Similar presentations


Ads by Google