©1996-2011 All Rights Reserved. Online
Journal of Bioinformatics . You may not store these pages in any form except for your
own
personal use. All other usage or distribute on is illegal under
international
copyright treaties. Permission to use any of these pages in any other
way
besides the before mentioned must be gained in
writing
from the publisher. This article is exclusively copyrighted in its
entirety to
OJB publications. This article may be copied once but may not be,
reproduced or re-transmitted
without the
express permission of the editors. This
journal
satisfies the refereeing requirements (DEST) for the Higher Education
Research Data Collection (Australia). Linking:To link to this page or any pages linking to this
page you
must link directly to this page only here rather than put up your own
page.
OJBTM
Online Journal of Bioinformatics ©
Volume 12(1):9-17, 2011
Prediction of mutagenicity
of compounds by Support Vector Machine
Anju Sharma1,2*, Rajnish Kumar1,2,
Pritish Varadwaj1
1Department
of Bioinformatics, Indian Institute of Information Technology Allahabad,Deoghat, Jhalwa, Allahabad-211012, Uttar Pradesh, India. 2Amity
Institute
of Biotechnology (AIB), Amity University Uttar Pradesh (AUUP),
Lucknow-206010, Uttar Pradesh, India
ABSTRACT
Sharma A,
Kumar R, Varadwaj P. Prediction of mutagenicity of compounds by Support Vector
Machine, Online
J Bioinformatics, 12(1):9-17,
2011. Various computational methods have been developed for mutagenicity prediction for
in-vitro or in-vivo toxicity prediction. Radial Basis Function (RBF)
kernel based Support
Vector Machine (SVM) classification model was used for the prediction of mutagenicity using 17
physicochemical descriptors. The selection of optimal hyperplane
parameters were performed with 1696 training compound data and the
prediction
efficiency of proposed classifier were tested on remaining 566 test data.
The overall prediction efficiency was,
71.73%. Youden’s index and Matthew
correlation index
were found to be 0.43 and 0.43 respectively and the Area under Receiver
Operating Curve
(ROC) was found to be 0.7847. The overall performance of the model was
equivalent to other reported methods.