_PROBLEM CoEPrA-2006_Classification_004

_GROUP_NAME Artem Cherkasov

_GROUP_MEMBERS
Emre Karakoc
Cenk Sahinalp
Artem Cherkasov

_ADDRESS
University of British Columbia, Medicine
Simon Fraser University, Computer Science
Vancouver, BC, Canada

_MODELING_PROCEDURE
Descriptors:
Based on our previous experience with QSAR modeling of peptides and
General QSAR clustering and classification of bioactivity properties,
we
decided to use the following strategy.
First, we have optimized the geometry of the studied peptides using
MMFF94 force field; carboxylic groups have been deprotonated, amino
groups -
protonated, partial charges computed according to [1].
Then we have computed QSAR descriptors that describe an entire peptide
molecule (global parameters) as well as descriptors corresponding to
constituent aminoacids considered in the context of their peptide
environment (we did not use 'isolated aminoacids' approximation).
Thus, for all peptides in the testing and training sets of
Classification_004 problem, we initially calculated > 400 various 3D
and 2D QSAR parameters that included:

- 50 global 'inductive' QSAR descriptors as described in [2].
- 10 local 'inductive' QSAR descriptors (computed toward CA atom) have
been calculated for each aminoacid of a given 8-mer; therefore, 80
additional 'inductive' QSAR descriptors have been produced.
- 260 global atomic type-specific 'inductive' QSAR descriptors -
(previously unpublished parameters) that have been computed additively
For specific atomic types presented in the studied peptides;
- We have also computed ~90 conventional 3D and 2D global QSAR
parameters which are implemented within the MOE modeling package [3].

All 'inductive' QSAR descriptors that are described above, have been
calculated by our own SVL scripts for the MOE; most of them can be
freely
downloaded through the SVL exchange.

Modeling Procedure:
We used our linear optimization method based on a distance measure[4]
for calculating our prediction model. Given the calibration data-set 
with
our optimization approach aims to find a weighted Minkowski distance
that maximizes the difference between active and inactive peptides. The
seperation  between active and inactive compounds are written as a
linear program (LP) where all the descriptors are given to the program.
In order to eliminate the problem of the missing descriptors for some
compounds such as certain atom types these descriptors are removed.
The final set contains 414 descriptors and all the descriptors are
normalized.
Although all the descriptors are given to the LP formulation only 100
of them are used in the optimal distance measure. You can see these
descriptors in the attachment.

We trained our prediction model using whole calibration data-set and
the quality of our model is determined using the accuracy of the 
prediction
results  which is calculated using a modification of the k nearest
neighbor (kNN)
classification.
kNN based classification assigns the activity of an peptide, P, as the
majority of the k nearest neighbors of P using the distance model
determined by the LP optimization.
Instead of looking only k compounds we consider all the compounds and
calculate the average distance between P and active compounds as well 
as
the distance between P and inactive compounds.
The activity of the test compounds is determined as the miniminum of
these two distances.
For the test data we have:
Sensitivity:0.84
Specificity:0.97
Accuracy:0.95

[1] Cherkasov, A. Inductive Electronegativity Scale. Iterative
Calculation
of Inductive Partial Charges. Journal of Chemical Information and
Computer
Sciences, 2003, 43, 2039-2047. Cherkasov, A., Z. Shi, Y. Li, S.M.
Jones, M.
Fallahi, G.L. Hammond. 'Inductive' Charges on Atoms in Proteins:
Comparative
Docking with the Extended Steroid Benchmark Set and Discovery of a
Novel
SHBG Ligand. Journal of Chemical Information and Modelling, 2005, 45,
1842-1853.
[2] Cherkasov, A. 'Inductive' Descriptors. 10 Successful Years in QSAR.
Current Computer-Aided Drug Design, 2005, 1, 21-42.
[3] Molecular Operational Environment, 2005, by Chemical Computing
Group
Inc., Montreal, Canada.
[4] Karakoc E., Cherkasov A., Sahinalp S. C.  Distance Based Algorithms
for
Small Biomolecule Classification and Structural Similarity Search.
ISMB'06,
14th Annual International conference on Intelligent Systems for
Molecular
Biology, Fortaleza, Brazil 2006.
[5] SNNS: Stuttgart Neural Network Simulator; Version 4.0, University
of
Stuttgart, 1995.

_PREDICTION
Obj_00001 -1
Obj_00002 -1
Obj_00003 +1
Obj_00004 -1
Obj_00005 -1
Obj_00006 -1
Obj_00007 -1
Obj_00008 -1
Obj_00009 -1
Obj_00010 -1
Obj_00011 -1
Obj_00012 -1
Obj_00013 -1
Obj_00014 -1
Obj_00015 -1
Obj_00016 +1
Obj_00017 -1
Obj_00018 -1
Obj_00019 -1
Obj_00020 -1
Obj_00021 -1
Obj_00022 -1
Obj_00023 -1
Obj_00024 -1
Obj_00025 -1
Obj_00026 +1
Obj_00027 -1
Obj_00028 -1
Obj_00029 -1
Obj_00030 -1
Obj_00031 -1
Obj_00032 -1
Obj_00033 -1
Obj_00034 +1
Obj_00035 +1
Obj_00036 -1
Obj_00037 -1
Obj_00038 -1
Obj_00039 +1
Obj_00040 -1
Obj_00041 -1
Obj_00042 -1
Obj_00043 -1
Obj_00044 -1
Obj_00045 +1
Obj_00046 +1
Obj_00047 -1
Obj_00048 -1
Obj_00049 +1
Obj_00050 -1
Obj_00051 -1
Obj_00052 +1
Obj_00053 -1
Obj_00054 +1
Obj_00055 -1
Obj_00056 -1
Obj_00057 +1
Obj_00058 -1
Obj_00059 +1
Obj_00060 +1
Obj_00061 -1
Obj_00062 -1
Obj_00063 -1
Obj_00064 +1
Obj_00065 +1
Obj_00066 -1
Obj_00067 -1
Obj_00068 -1
Obj_00069 -1
Obj_00070 -1
Obj_00071 -1
Obj_00072 -1
Obj_00073 -1
Obj_00074 -1
Obj_00075 -1
Obj_00076 -1
Obj_00077 +1
Obj_00078 -1
Obj_00079 -1
Obj_00080 +1
Obj_00081 -1
Obj_00082 -1
Obj_00083 -1
Obj_00084 +1
Obj_00085 -1
Obj_00086 -1
Obj_00087 -1
Obj_00088 -1
Obj_00089 -1
Obj_00090 -1
Obj_00091 -1
Obj_00092 -1
Obj_00093 +1
Obj_00094 -1
Obj_00095 -1
Obj_00096 -1
Obj_00097 -1
Obj_00098 -1
Obj_00099 -1
Obj_00100 +1
Obj_00101 +1
Obj_00102 -1
Obj_00103 -1
Obj_00104 -1
Obj_00105 -1
Obj_00106 -1
Obj_00107 -1
Obj_00108 -1
Obj_00109 -1
Obj_00110 -1
Obj_00111 -1