Table Of Contentfi
Arti cial Neural Network
for Drug Design, Delivery
and Disposition
Edited by
Munish Puri
Yashwant Pathak
Vijay Kumar Sutariya
Srinivas Tipparaju
Wilfrido Moreno
AMSTERDAM(cid:1)BOSTON(cid:1)HEIDELBERG(cid:1)LONDON
NEWYORK(cid:1)OXFORD(cid:1)PARIS(cid:1)SANDIEGO
SANFRANCISCO(cid:1)SINGAPORE(cid:1)SYDNEY(cid:1)TOKYO
AcademicPressisanimprintofElsevier
AcademicPressisanimprintofElsevier
125LondonWall,LondonEC2Y5AS,UK
525BStreet,Suite1800,SanDiego,CA92101-4495,USA
225WymanStreet,Waltham,MA02451,USA
TheBoulevard,LangfordLane,Kidlington,OxfordOX51GB,UK
Copyright©2016ElsevierInc.Allrightsreserved.
Nopartofthispublicationmaybereproducedortransmittedinanyformorbyanymeans,electronicor
mechanical,includingphotocopying,recording,oranyinformationstorageandretrievalsystem,without
permissioninwritingfromthepublisher.Detailsonhowtoseekpermission,furtherinformationaboutthe
Publisher’spermissionspoliciesandourarrangementswithorganizationssuchastheCopyrightClearance
CenterandtheCopyrightLicensingAgency,canbefoundatourwebsite:www.elsevier.com/permissions.
ThisbookandtheindividualcontributionscontainedinitareprotectedundercopyrightbythePublisher
(otherthanasmaybenotedherein).
Notices
Knowledgeandbestpracticeinthisfieldareconstantlychanging.Asnewresearchandexperiencebroaden
ourunderstanding,changesinresearchmethods,professionalpractices,ormedicaltreatmentmaybecome
necessary.
Practitionersandresearchersmustalwaysrelyontheirownexperienceandknowledgeinevaluatingandusing
anyinformation,methods,compounds,orexperimentsdescribedherein.Inusingsuchinformationormethods
theyshouldbemindfuloftheirownsafetyandthesafetyofothers,includingpartiesforwhomtheyhavea
professionalresponsibility.
Tothefullestextentofthelaw,neitherthePublishernortheauthors,contributors,oreditors,assumeanyliability
foranyinjuryand/ordamagetopersonsorpropertyasamatterofproductsliability,negligenceorotherwise,
orfromanyuseoroperationofanymethods,products,instructions,orideascontainedinthematerialherein.
ISBN:978-0-12-801559-9
BritishLibraryCataloguing-in-PublicationData
AcataloguerecordforthisbookisavailablefromtheBritishLibrary
LibraryofCongressCataloging-in-PublicationData
AcatalogrecordforthisbookisavailablefromtheLibraryofCongress
ForinformationonallAcademicPresspublications
visitourwebsiteathttp://store.elsevier.com/
TypesetbyTNQBooksandJournals
www.tnq.co.in
PrintedandboundintheUnitedStatesofAmerica
Dedication
Thisworkisdedicatedtomybigfamily,withouttheirpatienceandincredible
constantsupportitwouldn’tbepossibletodeliverthisbook.Theyalldeserve
recognition for shapingmy life andgivingme time.
MunishPuri, MS, PhD
To the loving memories of my parents memories of his parents, Dr Keshav
Baliram Hedgewar, who gave proper direction, my beloved wife Seema, who
gave positive meaning, and my son Sarvadaman, who gave a golden lining
tomy life.
Yashwant Pathak, PhD
This work isdedicated to theloving memoriesof my fatherwho passed away
on April 22, 2013.
Vijay Kumar Sutariya, PhD
I dedicate this book to my family, friends, and teachers and everybody who
made an impact inmy life.
MyspecialacknowledgmentgoestomywifeKiranandchildrenShraddhaand
Krishna, without their encouragement this would not havebeen possible.
Parentsareourfirstteachers,andIamextremelyfortunatetohaveparentsthat
taught me alot in my life. I would like to dedicate this tothem.
Srinivas Tipparaju, PhD
TotheIbero-AmericanScienceandTechnologyEducationConsortium(ISTEC)
for giving us the platform and opportunity to meet and start a journey to
improve education throughout the Ibero-American region. To our wives and
offspring that have supported us throughout this journey that we not only
have passion for but also are fully committed to.
Luis Fernando Cruz, PhD
and
WilfridoMoreno, PhD
Contributors
Snezana Agatonovic-Kustrin MARA University of Technology Selangor,
Malaysia
Orhan E. Arslan Department of Pathology and Cell Biology, University of
South Florida Morsani Collegeof Medicine,Tampa,FL, USA
PruthviRajBejugam NationalCentreforCellScience,NCCSComplex,Pune
University Campus, Pune,India
Jonathan Bernick Independent Consultant, Omaha, NE
MarilynBui PathologyandCellBiology,UniversityofSouthFlorida,Tampa,
FL, USA; AnalyticMicroscopy, Moffitt Cancer Center, Tampa, FL,USA
Jeffrey Burgess Department of Pharmaceutical Sciences, College of
Pharmacy, University of South Florida,Tampa,FL, USA
Julio Caballero Centro de Bioinformatica y Simulacion Molecular,
Universidad de Talca,Talca, Chile
Tapash Chakraborty Department of Pharmaceutical Sciences, Dibrugarh
University, Dibrugarh, India
Sharmistha P. Chatterjee Engineering Technology & Computer Science,
Broward CollegeNorth Campus, Hindu University ofAmerica,
Lighthouse Point, FL, USA;Department ofComputer & Electrical
EngineeringandComputerScience,FloridaAtlanticUniversity,BocaRaton,
FL, USA
Harsh Chauhan School of Pharmacy and Health Professions, Creighton
University, Omaha,NE
Luis Fernando Cruz Quiroga Complex Systems & Education Network
for the Ibero-American Science and Technology Education Consortium
(SCED-ISTEC)
MalayK.Das DepartmentofPharmaceuticalSciences,DibrugarhUniversity,
Dibrugarh, India
Pranab Jyoti Das Department of Pharmaceutical Sciences, Dibrugarh
University, Assam, India
Todd Daviau CoreRx,Inc., Clearwater, FL,USA
Meng Joo Er School of Electrical and Electronic Engineering, Nanyang
Technological University, Singapore
Michael Fernandez Virtual Nanoscience Laboratory, CSIRO Materials
Science & Engineering, Parkville, VIC, Australia
AnastasiaGroshev UniversityofSouthFlorida,MorsaniCollegeofMedicine,
Tampa,FL, USA
xv
xvi Contributors
Manish K. Gupta School of Pharmacy, Lloyd Institute of Management and
Technology, Greater Noida,UttarPradesh, India
Swati Gupta School of Pharmaceutical Sciences, Apeejay Stya University,
Gurgaon, Haryana,India
Syeda Saba Kareem Pharmacy Department, St. Joseph’s Hospital, Tampa,
FL, USA
MarkLloyd Analytic Microscopy, MoffittCancer Center, Tampa, FL, USA
Matthew MacPherson Department of Chemical Engineering, College of
Engineering,University of South Florida,Tampa,FL, USA
Vineetha Mandlik National Centre for Cell Science, NCCS Complex,
Pune University Campus, Pune, India
Vijay Masand Department of Chemistry, Vidya Bharti College, Amravati,
Maharashtra
Bhaskar Mazumder Department of Pharmaceutical Sciences, Dibrugarh
University, Assam, India
Brain McMillan CoreRx,Inc., Clearwater, FL,USA
WilfridoAlejandroMoreno DepartmentofElectricalEngineering,University
of South Florida, Tampa,FL,USA; R&D of Ibero-American Scienceand
Technology Education Consortium (ISTEC)
DavidMorton SchoolofPharmacyandAppliedScience,LaTrobeInstituteof
Molecular Sciences,La Trobe University, Bendigo,VIC, Australia
Timothy Padawer Department of Pharmaceutical Sciences, College of
Pharmacy,University of South Florida, Tampa,FL, USA
Abhijit S. Pandya Department of Computer & Electrical Engineering and
Computer Science, Florida AtlanticUniversity, Boca Raton, FL,USA
Jayvadan Patel Nootan Pharmacy College, S.K. Patel Campus, Visnagar,
Gujarat, India
AnitaPatel NootanPharmacyCollege,S.K.PatelCampus,Visnagar,Gujarat,
India
Yashwant Pathak USF College of Pharmacy, University of South Florida,
Tampa,FL,USA
Dev Prasad Formulation development,FreseniusKabiUSA, Skokie, IL
Charles Preuss College ofMedicine, University of South Florida, FL, USA
Munish Puri Electrical Engineering, University of South Florida, Tampa, FL,
USA; AnalyticMicroscopy,Moffitt Cancer Center, Tampa,FL,USA;
Visiting Fellow,National CancerInstitute, NIH, Bethesda, MD, USA
RavindraK.Rawal DepartmentofPharmaceuticalChemistry,ISFCollegeof
Pharmacy,Moga,Punjab, India
Shailza Singh National Centre for Cell Science, NCCS Complex, Pune
University Campus, Pune, India
AumSolanki DepartmentofPharmaceuticalSciences,CollegeofPharmacy,
University of South Florida, Tampa,FL,USA
Contributors xvii
Pochi R. Subbarayan Department of Medicine, Division of Hematology
and Oncology,University ofMiamiMiller SchoolofMedicine, Miami,
FL, USA
Srinivas M. Tipparaju Department of Pharmaceutical Sciences, College of
Pharmacy, University of South Florida,Tampa,FL, USA
Yong Zhang School of Electrical and Electronic Engineering, Nanyang
Technological University, Singapore
Foreword
The drug discovery research and development process (R&D) is complex. A
successful drug discovery requires the best scientific minds and expertise
focusing on resource and task management under significant time and cost
constraints. From a system design point of view, complexity arises from the
considerable underlying uncertainties and the wide range of dependent and
independentvariables.Considerthatthepotentialtargetforanewdrugdesign
mustbediscernedfromadatabaseof20,000to25,000humangenesmadeup
ofthree billionindividualbasepairstomatchtheright bindingpocketofthe
target protein.
The drug discovery R&D process is a multiobjective parallel challenge at both
theinsilicoandtheexperimentallevels.Thecapabilitiesofavailablecomputa-
tionaltoolshavebeenrealizedintheemergingareasofcombinatorialchemis-
try and high-throughput screening to handle large data of over 35 million
chemical compounds and their probable physical, chemical, and structural
properties. Improvements in computational capabilitiesdsuch as those due
to the application of artificial neural networks (ANNs)dare continuing to
work their way into the state of the art and are positively contributing to the
challenges of addressing complex, multiobjective optimization and focused
searches in uncertain and complex environments. The application of ANNs
tothe drugdiscoveryR&D process is the maintopic ofthis book.
ANNs are computational learning machine networksdinspired by human
brain neuronsdthat utilize nonlinear mapping techniques. As is evident
from the chapters in this book, ANNs are very suitable for application to
drug discovery R&D. They employ highly parallel processing techniques that
can address complex system environments characterized by a high degree of
uncertainty over a wide range of independent variables and many dependent
variables and can be used as a predictive tool as they learn from past experi-
ences and adapt. The ANN provides a pathway to tackle significant time and
cost constraints due to its ability to address the nonlinear and nonparametric
nature ofthe problems.
xix
xx Foreword
Computational model developers, researchers, medicinal chemists, and the
many other experts that contribute to the highly sequential process of drug
discoveryanddevelopmentwillfindthistextparticularlyuseful.Aspresented
herein19chapterspreparedbymanyoftheworld’sleadingexperts,theANN
plays acentralrole in drug design, discovery,delivery,anddisposition.
The19 chapters are assembled intofive main sections:
Section I: Basicsof ANN: ConceptandStrategyin Drug Design
Section II: Basics and Application ofANN in Drug Discovery
Section III: ANN in Drug Delivery
Section IV:ANN in DrugDisposition
Section V:ANN inVarious Applications in Medicine
These chapters include, but are not limitedto, coverageof basicconcepts and
modeling, the role of ANNs in target validation, genetic algorithm optimiza-
tionin drug design, neurobiological computation, challenges in Chemoinfor-
matics, the impact of ANN in quantitative structureeactivity relationship and
in computational modeling, data mining in drug discovery and design, drug
transportmodelingandsimulation,drugformulationanddrugadministration
strategies, pharmaceutical product and process development, computational
complexity, adaptive modeling and intelligent control, and cancer detection
and treatment.
I would like to congratulate the Editors for preparing this outstanding multi-
disciplinary text that bridges the space between engineering and health
sciences.Dr.MorenoandDr.PuridrepresentingengineeringdandDr.Sutariya,
Dr. Tipparaju, and Dr. Pathakdrepresenting the health sciencesdhave
worked together to create what is sure to become a standard reference for
drug discovery R&D researchers, students, and professionals.
I expect that the scientificcommunity worldwidewill welcome this effort!
Dr. Robert H.Bishop, P.E.
Dean,College ofEngineering
The University ofSouth Florida
Preface
Moderndrug discovery is anoutcome ofcollaborativeand cooperativeefforts at
thelevelofresearchersinacademic,industry,andgovernmentresearchinstitutions.
Computational processing and molecular modeling help scientists to harness
their knowledge gained from recent advances in genomics and proteomics to
understandbiologicalsystemsanddisordersaswellasdiseases.
The research and development (R&D) process is a complexand challenging task
thatinvolvesresourceandtaskmanagement,thebestscientificmindsandexper-
tise, time factors, and cost. The complexity of the biological system starts from
the potential target for a new drug design from the database of 20,000 to
25,000 human genes made up of three billion individual base pairs to match
a right binding pocket of the target protein. Target validation is a complex and
crucial step in drug development that helps scientist to avoid any frustrating
dead end research pathways. Medicinal chemists optimize the lead compound
to become a potential drug by understanding the structural parameters of the
target. The role of technology, computational tools, and smart algorithms is
verycrucialattheseearlier stagesofdrugdevelopment.Anymistakesandwrong
assessmentsinprioritizingtheleadcompoundmayaffectcost,time,andresearch
efforts.
Artificialneuralnetworks(ANNs)arewidelyusedforvariousbiomedicalapplica-
tions like computational chemistry at the molecular level, bioinformatics,
Chemoinformatics,andquantitativestructureeactivityrelationships(QSARs).Books
availableonANNexpressthetheoreticalortechnicalaspectsofthemathematical
modeling involved in the ANN approach to solve a problem in Chemoinfor-
matics.TheuseofANNinmedicinalchemistryisquitecommon.Computer-aided
designs of drugs use computational methods to design ligands and structure-
baseddrugdesigns.Basedonthebindingpocketaffinityofthetargetanddatabase
techniques,theoptimizationofdesignparametersarepredictedinmathematical
modeling.Computationalmethodsareusedtounderstanddiseasesatthemolec-
ularlevelandtodesignasafeandeffectivedrug.Molecularmechanicsareahelp-
ingtooltopredicttheconformalchangesintarget-basedquantitativemodelsand
binding affinity in the whole design process. Computational models are often
xxi
xxii Preface
structured around the ligandeprotein interaction and the target’s structural
parameters-based analysis.
It takes 8e15 years todevelopanew drugfrom the time it isdiscovered (dis-
coverytimelessthan1year)tomakingitavailableinthemarket.Theimpor-
tantfactorotherthantimeisthecostinvolvedintheR&Dofdrugdevelopment
(including experimental failures), estimated around $800m to $1b. The
mounting cost of phase II and phase III trials and reducing attrition rates are
additional challenges for the pharmaceutical industry. For every 5000 to
10,000compounds in the R&D process,only one receives FDAapproval.
ANN is widely used in QSAR in situations when the dataset is very large and
cannot solve with linear functions. Multilayers are designed as hidden neuron
layers with a varying set of neuron numbers. Input variables are selected from
theinformationrelatedtothedrug’sparameterslikeconcentration,compounds,
ligands,etc.andprocessedthroughmultipletrainingstepstomeettheoutput.
The predicted value is analyzed and compared with the known value. The
difference in predicted and known values is propagated backward until the
differencebecomesnegligiblysmalltoachievetheestimateddrugvalues.
Based on the above discussion we believe that there is a need for a reference
book that will address various aspects of ANN and its applications in drug
design, delivery, and dispositions. After carefully studying the literature we
foundthatthereareseveralbooksavailableonthemarketrelatedtocomputa-
tionaldrugdesign,molecularmodeling,Chemoinformaticsonthedrugdesign
side,andANNapplicationsinbiomedical,cancer,cardiovascular,andmathe-
maticalmodelinginneuroscience.However,thereisnoconsolidatedreference
book that discusses the applications of ANN in drug design, delivery, and
disposition.
This interface of ANN from the computational engineering side and drug
design, discovery, delivery, and disposition from the medicine side will not
onlysolvetheproblemofan8e15years-longperiodofdrugdesignanddevel-
opmentbutalsowillgiveaparadigmshiftindesigningnewmodelsandhelpin
designing a predictive tool for an effective drug development and disposition
system.
ANN’s big advantage of learning and self-correcting ability even in a highly
nonlinear, complex, noisy environment will be a milestone in the direction
ofdrugdesigningandcontrolleddeliveryforthefuturepharmaceuticalindus-
try.TheANNcanpredicttodeliverthedrugevenindeepbrainareaswiththe
help of implants in Parkinson’s and epilepsy diseases. Our book will be a
unique, knowledgeable resource for the researchers, scientists, and academics
working in the medicine and computational modeling communities and will
be atrendsetterin this field.
Description:Artificial Neural Network for Drug Design, Delivery and Disposition provides an in-depth look at the use of artificial neural networks (ANN) in pharmaceutical research. With its ability to learn and self-correct in a highly complex environment, this predictive tool has tremendous potential to help r