Predicting Structural Features

1
Predicting Structural Features

• Chapter 12

2
Structural Features

• Phosphorylation sites
• Transmembrane helices
• Protein flexibility

3
Accuracy Measures Revisited

• Level
• Individual residues
• Complete helix or strand

4
Residue-Level Measures

• Q3
• Percentage of residues predicted correctly
• If one state (eg, Coil) is very common (eg, 50),
  blind guessing can give a large Q3!
• Matthews correlation coefficient
• C (TPxTN - FNxFP)/v(TPFP)(TPFN)(TNFP)(TNFN)
• Defined for each state
• More balanced than Q3 in range 1
• Random prediction C 0

5
Structural Element-Level Measures

• SOV
• based on the overlap of predicted segments of
  helix, strand etc. with the observed segments of
  the same type
• The N-score
• specialized for transmembrane protein predictors
• Should TMHMM2 be changed? Should your model?

6
Predicting Helices

• Residue propensities
• score for a given structure class for each
  residue, a
• P(H a) is proportional to P(a H) / P(a)
• Why? Bayes Rule is your friend!
• P(H a) P(a H)P(H) / P(a)
• P(H) doesnt depend on a, so
• P(H a) proportional to P(a H) / P(a)

Can this be used to see how to group helix states?
7
Identical short segments rarely fold differently

• Local sequence is highly important to secondary
  structure.
• But, this sequence occurs in two proteins and
  takes very different forms
• KGVVPQLVK
• There is significant information about structure
  in local sequence.

8
I-sites Sequence Database

• About 250 short segments (3-19 residues) that
  show strong correlation between sequence and
  structure
• Example shows
• phi and psi angles, log-odds matrix
• superimposed backbones
• representative structure

9
Nearest Neighbor Prediction Methods

• Predict secondary structure based on
• Local alignments of the query sequence to a
  database of sequences of known structure
• Alignment score functions are often
  special-purpose, and may include helix/sheet/coil
  propensity information
• Homologous sequences are often included in the
  database
• Prediction based on weighted votes of nearest
  neighbors (usually only central residue of
  alignment is predicted)
• 73.5 Accuracy (Q3)

10
A different application prediction of misfolding

• Diseases such as Alzheimers involve protein
  misfolding.
• Usually, the misfolded region ends up as
  Beta-strands.
• How could we use secondary structure information
  to predict which proteins will potentially
  misfold?

11
H?PHidden Beta Propensity

• Key idea Tertiary contacts (TC)
• TC is number of contacts a residue has with
  others at least 4 residues away
• Alpha helices tend to be in regions of HIGH TC
• Beta strands tend to be in regions of LOW TC
• Look for query residues whose nearest neighbors
  are strange with respect to TC and alpha/beta
  state
• Low TC regions with lots of Alphas
• High TC regions with lots of Betas
• Performance results?

12
Neural Nets

• Each node computes a simple function of its
  inputs.
• The weighted sum of the inputs are added to a
  bias term and squashed
• I ? w??-1
• ??(I?)
• The output, ?, is then propagated to nodes in the
  next layer.

13
Training Neural Nets

• Back-propagation
• Optimizes the weights and bias terms
• Minimize the error function (difference between
  predicted and observed)
• RMS
• Relative Entropy
• Iterative process
• Final weights shown for a secondary structure NN
  alpha helix output layer.
• Over-fitting can be reduced by training for fewer
  iterations

14
Adaptive Encoding and Weight Sharing

• Orthogonal encoding
• Each residue feeds three hidden nodes
• The weights for all red nodes are tied together
• Each group of three nodes learns the same
  encoding of the 20 amino acids

15
Engineering Intuition Into NNs

• Alpha helices have a period of 3.6 residues per
  turn
• A NN can be specially designed to reflect that
• Using this, plus adaptive encoding
• Q3 66
• Adding homology Q3 73

16
HMMs and Transmembrane Proteins (again)
17
HMMTOP Architecture

• TMHs 17-25 residues
• Tails 1-15 residues
• Blue letters show structural state labels

18
TMHMM Architecture

• Helices are 5-25 residues
• Caps follow helices
• Cytoplasmic
• Loop 0-20 residues
• Globular 1 state
• Extra-cellular
• Long loop 0-100 residues
• Globular 3 states

19
Predicting Globular Proteins with Hidden Neural
Networks

• YASPIN
• Neural net predicts seven classes (He,H,
  Hb,C,Ee,E,Eb) using 15-residue window of PSSM
  input
• HMM filters this output
• Can you imagine how this is done?

20
Coiled-coil HMMMARCOIL
Design lets you start and end in any phase of the
heptad repeat
21
Support Vector Machines SVMs

• Classifiers
• Basic machine is a 2-class classifier
• Training Data
• set of labeled vectors
• ltx1, x2, ,xn, Cgt,
• Class C1 or C-1
• Supervised learning (like neural nets)
• Learn from positive and negative examples
• Output
• Function predicting class of unlabeled vectors

22
SVM Example

• Alpha helix predictor
• 15 residue window
• 21 numbers per residue
• Psi-BLAST PSSM 20 numbers
• spacer flag indicating off end of protein
• 315 numbers total per window
• Training samples
• Non-helix samples ltx1, x2, , x315, -1gt
• Helix samples ltx1, x2, , x315, 1gt
• Training finds function of X that best separates
  the non-helix from the helix samples

23
SVM vs NNas Classifiers

• Similarities
• Compute a function on their inputs
• Trained to minimize error
• Differences
• NNs find any hyperplane that separates the two
  clases
• SVMs find the maximum- margin hyperplane
• NNs can be engineered by designing their topology
• SVMs can be tailored by designing the kernel
  function

24
SVM Details
Separating Hyperplanes
Choose w, b to minimize w Subject to
Dual form (support vectors)
Kernel trick replace dot products by a
non-linear kernel bunction.
s.t.
where
25
Dubious Statement

• In marked contrast to NN, SVMs have few explicit
  parameters to fit
• The vector of weights, w, is as long as the
  number of training samples
• But the minimum-margin hyperplane will have most
  of the weights equal to zero only the support
  vectors will have non-zero weights.