Prediction of Antifungal Activity of Gemini Imidazolium Compounds

Research ArticlePrediction of Antifungal Activity of GeminiImidazolium Compounds

Aukasz PaBkowski1 Jerzy BBaszczyNski2 Andrzej Skrzypczak3

Jan BBaszczak3 Alicja Nowaczyk4 Joanna Wroacuteblewska5 Sylwia Kohuszko5

Eugenia Gospodarek5 Roman SBowiNski26 and Jerzy KrysiNski1

1Department of Pharmaceutical Technology Nicolaus Copernicus University Jurasza 2 85-094 Bydgoszcz Poland2Institute of Computing Science Poznan University of Technology Piotrowo 2 60-965 Poznan Poland3Institute of Chemical Technology Poznan University of Technology Skłodowskiej-Curie 2 60-965 Poznan Poland4Department of Organic Chemistry Nicolaus Copernicus University Jurasza 2 85-094 Bydgoszcz Poland5Department of Microbiology Nicolaus Copernicus University Skłodowskiej-Curie 9 85-094 Bydgoszcz Poland6Systems Research Institute Polish Academy of Sciences Newelska 6 01-447 Warsaw Poland

Correspondence should be addressed to Łukasz Pałkowski lukaszpalkowskicmumkpl

Received 21 October 2014 Revised 13 December 2014 Accepted 23 December 2014

Academic Editor Lei Chen

Copyright copy 2015 Łukasz Pałkowski et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

The progress of antimicrobial therapy contributes to the development of strains of fungi resistant to antimicrobial drugs Sincecationic surfactants have been described as good antifungals we present a SAR study of a novel homologous series of 140 bis-quaternary imidazolium chlorides and analyze them with respect to their biological activity against Candida albicans as one ofthe major opportunistic pathogens causing a wide spectrum of diseases in human beings We characterize a set of features ofthese compounds concerning their structure molecular descriptors and surface active properties SAR study was conducted withthe help of the Dominance-Based Rough Set Approach (DRSA) which involves identification of relevant features and relevantcombinations of features being in strong relationship with a high antifungal activity of the compounds The SAR study showsmoreover that the antifungal activity is dependent on the type of substituents and their position at the chloride moiety as wellas on the surface active properties of the compounds We also show that molecular descriptors MlogP HOMO-LUMO gap totalstructure connectivity index and Wiener index may be useful in prediction of antifungal activity of new chemical compounds

1 Introduction

In recent years the number of applications of quaternaryammonium compounds (QACs) has increased considerablyGemini QACs are a group of cationic surfactants containingtwo head groups and two aliphatic chains linked by a spacergroup

Practical implementation of gemini QACs is a resultof their surface active antielectrostatic and antimicrobialproperties

It has been demonstrated that gemini QACs exhibitproperties superior to mono QACs such as better solubilityhigher adsorption efficiency and better wetting and foaming

[1ndash4] Gemini QACs are more efficient in lowering surfacetension and have much lower critical micelle concentration(CMC) [5] Due to their higher surface activity they haveexcellent dispersion stabilization and soil clean-up properties[6 7] It has been also demonstrated that gemini QACs havegood antifungal activity [8ndash10] which is higher than monoQACs [11 12] So it is worth developing new more effectivecompounds such as gemini QACs

Because of the increasing resistance of microorganismsto commonly used disinfectants the synthesis of new typesof microbicides is a very important topic [13] Formationof resistant strains of fungi is not as common as formationof resistant strains of bacteria [14] Nevertheless knowledge

Hindawi Publishing CorporationBioMed Research InternationalVolume 2015 Article ID 392326 10 pageshttpdxdoiorg1011552015392326

2 BioMed Research International

of properties of chemical compound which influence theantifungal activity of gemini QACs enables designing andsynthesis of new active chemical entities

Themain goal of our studywas to investigate relationshipsbetween selected molecular parameters and features describ-ing chemical structure and surface active properties andantifungal activity (described as MFC (minimal fungicidalconcentration)) In MFC study Candida albicans ATCC90028 strainwas used In structure-activity relationship study(SAR) modified method based on a rough set theory wasemployed

Candida albicans is one ofmajor opportunistic pathogenscausing a wide spectrum of diseases in human beings It cancause infections that range from superficial infections of theskin to life-threatening systemic infections [15] Given thelimited number of suitable and effective antifungal agentstogether with increasing drug resistance of the pathogens itis important that new classes of antifungals are discovered[16] Moreover better understanding of which features ofchemical compounds decide high antifungal activity mayprovide further information useful for the improvement ofantifungal action

Data that describe the analyzed series of gemini imi-dazolium chlorides can be seen as classification datawhere parameters characterizing structure and surface activeproperties as well as molecular parameters are conditionattributes (independent variables) and antifungal activityis represented by class labels assigned to chlorides by adecision attribute (dependent variable) Structure-activityrelationships can be discovered from these data by explainingthe class assignment in terms of condition attributes To thisend we applied the rough set concept [17] and its particularextension called Dominance-Based Rough Set Approach(DRSA) [18ndash21]

2 Materials and Methods

21 Gemini Imidazolium Chlorides We analyzed 10 homolo-gous series of gemini imidazolium chlorides with hydropho-bic chain ranging from CH

3to C16H33

and with the lengthof spacer from C

2to C12 Synthesis surface active proper-

ties and antimicrobial activity of a part of 140 331015840-(120572120596-dioxaalkyl)bis(1-alkylimidazolium) chlorides were describedearlier [22] Moreover we determined molecular descriptorsfor synthetized structures The antifungal activity was deter-mined by theMFC valuesThe final stage of our study was ananalysis of structure-activity relationships using DRSA [21]

22 Chemical Structure Chemical structure of chlorides wasdescribed by the following parameters (see Figure 1 andTable 1)

(i) 119899 number of carbon atoms in 119899-spacer(ii) 119877 number of carbon atoms in 119877-substituent

23 Surface Active Properties Surface active properties ofanalyzed chlorides were described by the following param-eters

Table 1 Numerical coding of the structure of analyzed chlorides

Code Condition attributes119899-spacer 119877-substituent

1 CH3

2 C2H5 C2H5

3 C3H7 C3H7

4 C4H9 C4H9

5 C5H11 C5H11

6 C6H13 C6H13

7 C7H15 C7H15

8 C8H17 C8H17

9 C9H19 C9H19

10 C10H21 C10H21

11 C11H23

12 C12H25 C12H25

14 C14H29

16 C16H33

RR

OO

N NN+ N+

2Clminusn

CH2

Figure 1 Chemical structure of analyzed compounds

(i) CMC critical micelle concentration (molL)(ii) 120574CMC value of surface tension at critical micelle

concentration (mNm)(iii) Γ times 106 (G) value of surface excess (molm2)(iv) 119860 times 10minus20 molecular area of a single particle (m2)(v) Δ119866ads free energy of adsorption ofmolecule (kJmol)

24 Molecular Parameters We also considered molecularparameters of analyzed compounds which were calculatedwith Dragon and Gaussian software Molecular descriptor isthe final result of a logic and mathematical procedure whichtransforms chemical information encoded within a symbolicrepresentation of a molecule into a useful number or a resultof a standardized experiment [23] Those parameters were

(i) MLOGP Moriguchi octanol-water partition coeffi-cient

(ii) Balaban index (BI) Narumi topological index (NTI)total structure connectivity index (TSC) Wienerindex (WI) numerical parameters characterizingcompoundsrsquo topology

(iii) HOMO highest occupied molecular orbital(iv) LUMO lowest unoccupied molecular orbital(v) HOMO-LUMO gap (HL gap) the energy difference

between the HOMO and LUMO

BioMed Research International 3

(vi) dipole (dip) electric dipole moment(vii) radius of gyration (ROG) the root mean square

distance of the entitiesrsquo parts from either its center ofgravity or a given axis

(viii) molecular weight (MW) of compounds

25 Antifungal Activity Candida albicans ATCC 90028 mi-croorganisms were used to evaluate antifungal activity ofcompounds by minimal fungicidal concentration (MFC)MFC determination method was presented in [22]

According to the value of MFC objects were sorted intothree decision classes

(i) class good good antifungal properties MFC le 0028mML

(ii) classmediummedium antifungal properties 0028 ltMFC lt 01mML

(iii) class weak weak antifungal properties MFC ge 01mML

Values of MFC for activity classes were determined onthe basis of antimicrobial activity of benzalkonium chlorideand didecyldimethylammonium chloride used as referenceantifungals

26 SARAnalysis Based onDRSAmdashDescription of theMethodDRSA assumes that the value sets of condition attributesare ordered and monotonically dependent on the order ofdecision classes DRSA proved to be an effective tool inanalysis of classification data which are partially inconsistent[24 25] In the context of this study inconsistencymeans thatbetween a pair of chlorides the first one has not worse surfaceactive and molecular properties than the other although thefirst one is assigned to a worse class of antifungal activitythan the other The rough set analysis of consistent andinconsistent chlorides prepares the ground for induction ofdecision rules The rules derived from data structured usingthe concept of the DRSA are monotonic which means thatthey have the following syntax

ldquoif ati(chloride) ge vali and atj(chloride) ge valj and sdot sdot sdotand atp(chloride) ge valp then chloride is assigned toat least a given classrdquoldquoif atk(chloride) le valk and atl(chloride) le vall and sdot sdot sdotand ats(chloride) le vals then chloride is assigned to atmost a given classrdquo

where ath is an ℎth condition attribute and valh is a thresholdvalue of this attribute which makes an elementary conditionath(chloride) ge valh or ath(chloride) le valh composing acondition part of a rule indicating assignment of a chlorideto at least (or at most) a given class (weak medium or good)respectively In the above syntax of the rules it is assumedthat value sets of all condition attributes are numerical andordered such that the greater the value the more likely it isthat the chloride has good antifungal activity analogouslyit is assumed that the smaller the value the more likely itis that the chloride has weak antifungal activity Attributes

ordered in this way are called gain-type Cost-type attributeshave value sets ordered in the opposite direction suchthat elementary conditions on these attributes have oppositerelation signs In case of gemini imidazoliumchlorides data itis not known a prioriwhether condition attributes are gain orcost attributes Therefore we proceeded as described in [26]each original attribute is considered in two copies with thefirst copy assumed to be gain-type and the second cost-typeThe applied transformation of data is noninvasive that is itdoes not bias the relationships identified between conditionattributes and the decision attribute Then an inductionalgorithm constructs decision rules involving elementaryconditions on one or both copies of particular attributes Forexample for a rule indicating the assignment of a chlorideto class good (at least good) the following elementaryconditions concerning attribute ati may appear

(i) uarrati(chloride) ge vali1(ii) darrati(chloride) le vali2(iii) uarrati(chloride) ge vali1 and darrati(chloride) le vali2 which

boils down to ati(chloride) isin [vali1 vali2] if vali1 levali2

where uarrati and darrati are gain-type and cost-type copies ofattribute ati respectively Note that this transformation ofattributes allows global and local monotonic relationshipsto be discovered between condition attributes and classassignment A monotonic relationship is global when it canbe expressed by a single elementary condition concerninggain-type or cost-type attribute Local monotonicity relation-ship requires conjunction of two elementary conditions ofdifferent types In case of assignment of a chloride to classgood we can have such a local monotonicity relationshipfor example when concentration of a surface active propertyis below a certain point the greater the value the betterthe assignment but after that point further increase mayhave a negative effect (ie the lower the value the better theassignment)

3 Results and Discussion

31 Information System Information system is the basis ofSAR analysis of the chemical compounds It includes a set ofobjects (in rows) described by a set of attributes (in columns)The set of attributes is composed of condition and decisionattributes In our case condition attributes describe surfaceactive properties molecular descriptors and structure (thelength on 119899 spacer and the length of 119877-chain) of analyzedchlorides The decision attribute concerns antifungal proper-ties of bis-quaternary imidazolium chlorides represented bysome limit values of MFC for Candida albicansATCC 90028A part of information system is presented in Table 2

32 Decision Rules Table 3 includes strong and relevantdecision rules obtained for good andweak classes of chloridespresented in Table 2 These are rules selected from the set ofall minimal decision rules induced from information tableprocessed by DRSA


Table2Apartof

inform

ationsyste

m(10fro

m140ob

jects)

Num

ber119899119877

lgCM

CgC

MC119866119860119866

ads

MLO

GP

BINTI

WI

MW

HOMO

LUMO

HLgap

dip

ROG

TSC

MIC

[mML]

Class

12

1215

619

275

52202

0175

1397

12712

5275

25236minus038777minus019852minus018925

1646

4908

028

16937

Weak

22

2223

601

271

54208

0711

1407

14099

5768


0103

5294

0266

3558

Weak

32

3238

598

269

56213

1216

1405

15485

6307


2314

5804

0254

3295

Weak

42

4241

574

265

58217

1697

1397

16871

6877


35474

6246

0243

1634

Weak

52

5249

555

261

60223

2157

1386

18257

7468


8628

6777

0234

0712

Weak

62

6258

534

257

62227

2599

1373

19644

8074

39266minus038544minus01864

1minus019903

12501

725

0226

0086

Medium

72

7265

512

253

64235

3025

1359

2103

8692


16201

7791

0218

0020

Goo

d8

28

272

489

249

66239

4349

1346

22416

9319


20472

8282

0211

0005

Goo

d9

29

281

475

245

68243

4748

1333

23803

9952


24499

8831

0205

0002

Goo

d10

210

292

453

241

70248

5136

132

2518

91059


29011

9333

0199

0017

Goo

d


Table3Decision

rules

Num

ber

Con

ditio

nattributes

Exam

ples

Streng

thCon

firmationmeasure

s119899119877minuslogC

MC120574CM

CΓsdot10

6Asdot10

20Δ119866

ads119872

log119875

MW

WI

HLg

apBI

TSC

NTI

Decision

classgood

1le10

le0218

5403857

07500

2le11

ge3836

4903500

07005

3leminus017317

ge2103

4203000

08000

4leminus015191

le0199

4203000

07142

5le11

le0199

3902785

07058

6ge6

ge40

669

leminus017314

3702642

06315

7ge7

ge37863

ge12

4234

02428

06315

8ge7

le483

ge12

4231

02214

06315

9ge7

ge277le694

2902071

066

6610

[711]

ge262

2902071

066

66Decision

classweak

11le6

ge248

2601857

09523

12le6

ge252

2501785

09523

13le6

le2814

2401714

09523

14le6

ge253

2401714

09523

15le6

le254

2301642

09523

16le6

ge256

2201571

09523

17le5

ge501

2201571

09523

18le5

ge252

2101500

09545

19le5

ge521

2001428

09523

20le5

le232

2001428

09523


We did not induce rules for class ldquomediumrdquo since theserules are not interesting from the viewpoint of SAR analysis(it is more important to know what are the features ofchlorides with definitely good or weak antimicrobial proper-ties) However the presence of chlorides from the ldquomediumrdquoclass is important in the rule induction process The ruleswith conclusion ldquogoodrdquo discriminate chlorides with ldquogoodrdquoantimicrobial properties from those chlorides which haveldquomediumrdquo or ldquoweakrdquo properties (analogously for rules withconclusion ldquoweakrdquo)

The decision rules provide guidelines for synthesis ofnew compounds with better antifungal properties The rulesare characterized by various parameters such as examples(ie number of objects covering a given rule) strength (iethe proportion of objects covered by premise that are alsocovered by conclusion) or confirmation (ie measure that isquantifying the degree to which premise provides evidencefor conclusion)

In Table 3 only attributes that were present in decisionrules are included

Rules are characterized by their strength defined as a ratioof the number of chloridesmatching the condition part of therule to the total number of chlorides in the sample Sets ofdecision rules which are essential for the analysis presentedin this work were induced from gemini imidazolium chlo-rides data which were collected in an information system Apart of the system can be seen in Table 2 These data weretransformed as described above and structured accordingto the DRSA The induction algorithm that was applied toconstruct rules is called VC-DomLEM [27] The algorithmwas implemented as a part of software package called jMAF(httpidsscsputpoznanplsite139html) based on the javaRough Set (jRS) library The sets of induced rules were usedto construct component classifiers in variable consistencybagging [28 29] Variable consistency bagging (VC-bagging)[29] was applied to increase the accuracy of results producedby VC-DomLEM

Both rule relevance and relevance of attribute whichare present in condition part of rules were estimated bymeasuring Bayesian confirmation as described in [30] Inthis process decision rules were constructed repetitively onbootstrap samples and tested with chlorides that were notincluded in the samples

In the ldquogoodrdquo class of antifungal activity strong rulessupported by a large number of objects were obtained Themost interesting rules are characterized by high confirmationmeasures In decision rules covering chlorides with goodactivity against Candida albicans chlorides with 119899-spacerlonger or equal to 6 atoms of carbon predominate We canalso observe that optimal length of 119877-substituent is from 7 to11 carbon atoms in a chain Moreover those rules emphasizethat 120574CMC is important from the point of view of assigningnew compounds into a good class of activity As it wasmentioned before we included molecular descriptors intoour SAR analysis Results are as follows Moriguchi octanol-water partition should be in the range [3836 694] theenergy difference between the HOMO and LUMO should beless than or equal tominus017314 Balaban index should be greaterthan or equal to 1242 Narumi topological index should be

greater than or equal to 21 and total structure connectivityindex should be less than or equal to 0218

When we consider assigning new chlorides into weakdecision class the length of 119899-spacer in compoundrsquos moietyshould be shorter or equal to 6 atoms of carbon We canalso observe that values of surface tension at critical micelleconcentration greater or equal to 501 values of surfaceexcess greater or equal to 248 and values of free energyof adsorption of molecule less than or equal to 232 areimportant when considering weak activity against Candidaalbicans strains Decision rules for weak class of chloridesinclude only one molecular descriptor Moriguchi octanol-water partition coefficient in contrast to good activity classwhich included all molecular descriptors besides Wienerindex

33 Attribute Relevance Results of estimation of predictiveconfirmation of all attributes (structure surface active andmolecular ones) in rules induced for class good and weak arepresented in Figures 2 and 3

Let us interpret a rule as a consequence relation ldquoif E thenHrdquo where 119864 denotes rule premise and119867 rule conclusion Forrule relevance the Bayesian confirmation measure quantifiesthe contribution of rule premise 119864 to correct classification ofunseen individuals Many Bayesian confirmation measureshave been described in the literature of which we used themeasure 119904(119867 119864) This approach allows clear interpretation interms of a difference of conditional probabilities involving119867 and E that is 119904(119867 119864) = Pr(119867 | 119864) minus Pr(119867 | not119864)where probability Pr(sdot) is estimated from the test samples ofchlorides For the relevance of single attributes the Bayesianconfirmation measure quantifies the degree to which thepresence of attribute ati in premise 119864 denoted by at

119894⊳ 119864

provides evidence for or against conclusion 119867 of the ruleHere we used again measure 119904(119867 at

119894⊳ 119864) which in this

case is defined as follows 119904(119867 at119894⊳ 119864) = Pr(119867 | at

119894⊳

119864)minusPr(119867 | at119894not ⊳ 119864) Consequently attributes present in the

premise of a rule that assigns chlorides correctly or attributesabsent from the condition part of a rule that assigns chloridesincorrectly are considered more relevant

We can observe that attributes Moriguchi octanol-waterpartition coefficient the length of119877 substituent andHOMO-LUMO gap are the most relevant when the good class ofactivity is considered On the other hand the most relevantattributes for weak decision class are the length of 119899-spacerBalaban index and LUMOparameterThese results show thatall three types of parameters structure surface active andmolecularmight be helpful in assigning new chemical entitiesto a specific class of antifungal activity

Chemical structure of gemini surfactants influences notonly their surface properties but also their antimicrobialactivity It has been widely accepted that optimal antimicro-bial activity can be obtained from 10 to 18 atoms of carbonin an aliphatic chain with an optimum of 12 to 16 atoms ofcarbon depending on a bacterial strain [31] An elongationof the hydrophobic chain increases antimicrobial activitybut only to a given limit after which activity decreases Itwas also observed that the lowest MFC values are specificfor medium-length hydrophobic substituents attached to a


000002004006

logC

MC

Gad

sRG

yrat

ion

HO

MO

LUM

Odi

pole

Bala

ban

Wie

ner

gCM

CM

WTS

CN

arum

iH

OM

O-L

UM

O g

ap

MLO

GP

minus002

minus004

minus006

A n G R

Con

firm

atio

n m

easu

reS

Figure 2 Predictive confirmation of attributes for class good

000

005

logC

MC

Gad

s

RGyr

atio

n

HO

MO

LUM

O

dipo

le

Bala

ban

Wie

ner

gCM

C

MW

TSC

Nar

umi

HO

MO

-LU

MO

gap

MLO

GP

minus005

minus010

A GR n

Con

firm

atio

n m

easu

reS

Figure 3 Predictive confirmation of attributes for class weak

quaternary atomof nitrogen [32] Similar observations can befound in [33] Specific properties of gemini compounds withthe above mentioned length of hydrophobic substituentsare related to their ability to form and coexist with smallspherical micelles and large aggregates Below this range onlymicelles are found while above this range only aggregates areobserved [34]

In this paper it was found that good antifungal activity fora group of analyzed gemini chlorides is related to 119899-spacerequal to or longer than 6 atoms of carbon Moreover wediscoveredmore features being in a strong relationship with agood antifungal activity regarding Candida albicans strainsThose are not only the length of substituents in a moiety butalso logCMC and 120574CMC Moriguchi octanol-water partitioncoefficient the energy difference between the HOMO andLUMO Balaban index Narumi topological index and totalstructure connectivity index Those parameters should betaken into consideration when one will plan synthesis of newgemini chloride with a high anti-Candida albicans activity

34 Results of Stratified Cross-Validation The model con-structed by VC-bagging with VC-DomLEM componentclassifiers showed good classification performance in 5-foldstratified cross-validation which was repeated 100 timesfor a better reproducibility of results First we consideredaccuracy of distinction between chlorides that have good

and not good (ie medium or weak) antifungal activityproperties In this case on the average 773 of chlorideswere correctly classified (819 were correctly classified ashaving good properties and 707were correctly classified ashaving not good properties) Second we checked distinctionbetween chlorides having weak and not weak (ie mediumor good) antifungal activity properties On the average 862of chlorides were correctly classified in this case (809 werecorrectly classified as having weak properties and 881 werecorrectly classified as having not weak properties)

4 Conclusions

Decision rules presented in this study show that numberof carbon atoms in 119899-spacer number of carbon atoms in119877-substituent MlogP HOMO-LUMO gap total structureconnectivity index and Wiener index have the most influ-ence on the increase of antifungal activity of 331015840-(120572120596-dioxaalkyl)bis(1-alkylimidazolium) chlorides On the otherhand number of carbon atoms in 119899-spacer value of surfaceexcess and Wiener index affected decreasing of antifungalactivity of studied gemini imidazolium chlorides Obtainedresults are directions for synthesis of new active molecules ofgemini imidazolium chlorides possessing strong antifungalaction DRSA is a valuable tool to conduct SAR analysis ofchemical compounds

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

Łukasz Pałkowski and Jerzy Krysinski wish to acknowledgefinancial support from the Nicolaus Copernicus UniversityGrant no 845

References

[1] Z C Miao and J W Ren ldquoPreparation of novel diester gem-ini quaternary ammonium salt cationic surfactantrdquo AdvancedMaterials Research vol 1004-1005 pp 699ndash702 2014

[2] Z Miao F Wang D Deng Y Zhang X Huo and LWang ldquoPreparation of novel gemini quaternary ammonium saltcationic surfactantrdquo Applied Mechanics and Materials vol 174-177 pp 1433ndash1436 2012

[3] M S Bakshi and K Singh ldquoSynergistic interactions in themixedmicelles of cationic gemini with zwitterionic surfactantsfluorescence and Krafft temperature studiesrdquo Journal of Colloidand Interface Science vol 287 no 1 pp 288ndash297 2005

[4] O Zheng and J-X Zhao ldquoSolubilization of pyrene in aqueousmicellar solutions of gemini surfactants C12-s-C12-2Brrdquo Journalof Colloid and Interface Science vol 300 no 2 pp 749ndash7542006

[5] J-H Fuhrhop and T Wang ldquoBolaamphiphilesrdquo ChemicalReviews vol 104 no 6 pp 2901ndash2937 2004

[6] P Conte A Agretto R Spaccini and A Piccolo ldquoSoil reme-diation humic acids as natural surfactants in the washings of


highly contaminated soilsrdquo Environmental Pollution vol 135no 3 pp 515ndash522 2005

[7] E Schnell D Touraud R Gick and W Kunz ldquoPropertiesof a new hydrotrope hydrophobic molecule and its potentialapplicationsrdquo International Journal of Cosmetic Science vol 30no 5 pp 347ndash351 2008

[8] J Zabielska-Matejuk ldquoAntifungal properties of new quaternaryammonium compounds in relation to their surface activityrdquoWood Science and Technology vol 39 no 3 pp 235ndash243 2005

[9] C K L Ng D Obando F Widmer L C Wright T CSorrell andKA Jolliffe ldquoCorrelation of antifungal activitywithfungal phospholipase inhibition using a series of bisquaternaryammonium saltsrdquo Journal of Medicinal Chemistry vol 49 no 2pp 811ndash816 2006

[10] M Tischer G Pradel K Ohlsen and U Holzgrabe ldquoQuater-nary ammonium salts and their antimicrobial potential targetsor nonspecific interactionsrdquo ChemMedChem vol 7 no 1 pp22ndash31 2012

[11] A Shirai T Sumitomo M Kurimoto H Maseda and HKourai ldquoThe mode of the antifungal activity of gemini-pyridinium salt against yeastrdquo Biocontrol Science vol 14 no 1pp 13ndash20 2009

[12] A Shirai S Ueta H Maseda H Kourai and T Omasa ldquoActionof reactive oxygen species in the antifungal mechanism ofgemini-pyridinium salts against yeastrdquo Biocontrol Science vol17 no 2 pp 77ndash82 2012

[13] K Hegstad S Langsrud B T Lunestad A A Scheie MSunde and S P Yazdankhah ldquoDoes the wide use of quaternaryammonium compounds enhance the selection and spreadof antimicrobial resistance and thus threaten our healthrdquoMicrobial Drug Resistance vol 16 no 2 pp 91ndash104 2010

[14] A M Carmona-Ribeiro and L D M Carrasco ldquoFungicidalassemblies and their mode of actionrdquo OA Biotechnology vol 2no 3 article 25 2013

[15] F L Mayer D Wilson and B Hube ldquoCandida albicanspathogenicity mechanismsrdquoVirulence vol 4 no 2 pp 119ndash1282013

[16] C Ferreira S Silva F Faria-Oliveira E Pinho M Henriquesand C Lucas ldquoCandida albicans virulence and drug-resistancerequires the O-acyltransferase Gup1prdquo BMC Microbiology vol10 article no 238 2010

[17] Z Pawlak Rough Sets Theoretical Aspects of Reasoning aboutData Kluwer Academic Publishers Dordrecht The Nether-lands 1991

[18] J Błaszczynski S Greco R Słowinski and M Szelag ldquoMono-tonic variable consistency rough set approachesrdquo InternationalJournal of Approximate Reasoning vol 50 no 7 pp 979ndash9992009

[19] S Greco BMatarazzo and R Słowinski ldquoRough sets theory formulticriteria decision analysisrdquoEuropean Journal of OperationalResearch vol 129 no 1 pp 1ndash47 2001

[20] R Słowinski S Greco and B Matarazzo ldquoRough sets indecision makingrdquo in Encyclopedia of Complexity and SystemsScience R A Meyers Ed pp 7753ndash7786 Springer New YorkNY USA 2009

[21] R Słowinski S Greco and B Matarazzo ldquoRough-set-baseddecision supportrdquo in Introductory Tutorials in Optimization andDecision Support Techniques E K Burke and G Kendall Edschapter 19 pp 557ndash609 Springer New York NY USA 2ndedition 2014

[22] Ł Pałkowski J Błaszczynski A Skrzypczak et al ldquoAntimicro-bial activity and SAR study of new gemini imidazolium-basedchloridesrdquo Chemical Biology amp Drug Design vol 83 no 3 pp278ndash288 2014

[23] R Todeschini andVConsonniHandbook ofMolecularDescrip-tors Wiley-VCH New York NY USA 2000

[24] R Słowinski J Stefanowski S Greco and BMatarazzo ldquoRoughset based processing of inconsistent information in decisionanalysisrdquo Control and Cybernetics vol 29 no 1 pp 378ndash4042000

[25] R Słowinski S Greco and B Matarazzo ldquoRough set analysis ofpreference-ordered datardquo in Rough Sets and Current Trends inComputing J J Alpigini J F Peters A Skowron andN ZhongEds vol 2475 of Lecture Notes in Computer Science pp 44ndash59Springer Berlin Germany 2002

[26] J Błaszczynski S Greco and R Słowinski ldquoInductive discoveryof laws using monotonic rulesrdquo Engineering Applications ofArtificial Intelligence vol 25 no 2 pp 284ndash294 2012

[27] J Błaszczynski R Słowinski and M Szeląg ldquoSequential cover-ing rule induction algorithm for variable consistency rough setapproachesrdquo Information Sciences vol 181 no 5 pp 987ndash10022011

[28] J Błaszczynski R Słowinski and J Stefanowski ldquoFeature set-based consistency sampling in bagging ensemblesrdquo in FromLocal Patterns To Global Models (LEGO) ECMLPKDD Work-shop pp 19ndash35 2009

[29] J Błaszczynski R Słowinski and J Stefanowski ldquoVariableconsistency bagging ensemblesrdquo in Transactions on Rough SetsXI vol 5946 of Lecture Notes in Computer Science pp 40ndash52Springer Berlin Germany 2010

[30] J Błaszczynski R Słowinski and R Susmaga ldquoRule-basedestimation of attribute relevancerdquo in Rough Sets and KnowledgeTechnology J T Yao S Ramanna G Wang and Z Suraj Edsvol 6954 of Lecture Notes in Computer Science pp 36ndash44Springer Berlin Germany 2011

[31] M Lukac I Lacko M Bukovsky et al ldquoSynthesis and antimi-crobial activity of a series of optically active quaternary ammo-nium salts derived from phenylalaninerdquo Central EuropeanJournal of Chemistry vol 8 no 1 pp 194ndash201 2010

[32] P Bunio and J Chlebicki ldquoNew sorbic-type quaternary ammo-nium single-chain and gemini polymerizable surfactants syn-thesis interfacial properties and anti-electrostatic activityrdquo Col-loids and Surfaces A Physicochemical and Engineering Aspectsvol 413 pp 119ndash124 2012

[33] L Caillier E Taffin de Givenchy R Levy Y VandenbergheS Geribaldi and F Guittard ldquoPolymerizable semi-fluorinatedgemini surfactants designed for antimicrobial materialsrdquo Jour-nal of Colloid and Interface Science vol 332 no 1 pp 201ndash2072009

[34] M Pisarcik M Dubnickova F Devınsky I Lacko and JSkvarla ldquoDynamic light scattering and electrokinetic potentialof bis(quarternary ammonium bromide) surfactant micelles asthe function of the alkyl chain lengthrdquo Colloids and Surfaces APhysicochemical and Engineering Aspects vol 143 no 1 pp 69ndash75 1998

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Anatomy Research International

PeptidesInternational Journal of


Hindawi Publishing Corporation httpwwwhindawicom

International Journal of

Volume 2014

Zoology


Molecular Biology International

GenomicsInternational Journal of


The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014


BioinformaticsAdvances in

Marine BiologyJournal of



Signal TransductionJournal of


BioMed Research International

Evolutionary BiologyInternational Journal of



Biochemistry Research International

ArchaeaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Genetics Research International


Advances in

Virolog y

Hindawi Publishing Corporationhttpwwwhindawicom

Nucleic AcidsJournal of

Volume 2014

Stem CellsInternational



Enzyme Research



Microbiology


of properties of chemical compound which influence theantifungal activity of gemini QACs enables designing andsynthesis of new active chemical entities

Themain goal of our studywas to investigate relationshipsbetween selected molecular parameters and features describ-ing chemical structure and surface active properties andantifungal activity (described as MFC (minimal fungicidalconcentration)) In MFC study Candida albicans ATCC90028 strainwas used In structure-activity relationship study(SAR) modified method based on a rough set theory wasemployed

Candida albicans is one ofmajor opportunistic pathogenscausing a wide spectrum of diseases in human beings It cancause infections that range from superficial infections of theskin to life-threatening systemic infections [15] Given thelimited number of suitable and effective antifungal agentstogether with increasing drug resistance of the pathogens itis important that new classes of antifungals are discovered[16] Moreover better understanding of which features ofchemical compounds decide high antifungal activity mayprovide further information useful for the improvement ofantifungal action

Data that describe the analyzed series of gemini imi-dazolium chlorides can be seen as classification datawhere parameters characterizing structure and surface activeproperties as well as molecular parameters are conditionattributes (independent variables) and antifungal activityis represented by class labels assigned to chlorides by adecision attribute (dependent variable) Structure-activityrelationships can be discovered from these data by explainingthe class assignment in terms of condition attributes To thisend we applied the rough set concept [17] and its particularextension called Dominance-Based Rough Set Approach(DRSA) [18ndash21]

2 Materials and Methods

21 Gemini Imidazolium Chlorides We analyzed 10 homolo-gous series of gemini imidazolium chlorides with hydropho-bic chain ranging from CH

3to C16H33

and with the lengthof spacer from C

2to C12 Synthesis surface active proper-

ties and antimicrobial activity of a part of 140 331015840-(120572120596-dioxaalkyl)bis(1-alkylimidazolium) chlorides were describedearlier [22] Moreover we determined molecular descriptorsfor synthetized structures The antifungal activity was deter-mined by theMFC valuesThe final stage of our study was ananalysis of structure-activity relationships using DRSA [21]

22 Chemical Structure Chemical structure of chlorides wasdescribed by the following parameters (see Figure 1 andTable 1)

(i) 119899 number of carbon atoms in 119899-spacer(ii) 119877 number of carbon atoms in 119877-substituent

23 Surface Active Properties Surface active properties ofanalyzed chlorides were described by the following param-eters

Table 1 Numerical coding of the structure of analyzed chlorides

Code Condition attributes119899-spacer 119877-substituent

1 CH3

2 C2H5 C2H5

3 C3H7 C3H7

4 C4H9 C4H9

5 C5H11 C5H11

6 C6H13 C6H13

7 C7H15 C7H15

8 C8H17 C8H17

9 C9H19 C9H19

10 C10H21 C10H21

11 C11H23

12 C12H25 C12H25

14 C14H29

16 C16H33

RR

OO

N NN+ N+

2Clminusn

CH2

Figure 1 Chemical structure of analyzed compounds

(i) CMC critical micelle concentration (molL)(ii) 120574CMC value of surface tension at critical micelle

concentration (mNm)(iii) Γ times 106 (G) value of surface excess (molm2)(iv) 119860 times 10minus20 molecular area of a single particle (m2)(v) Δ119866ads free energy of adsorption ofmolecule (kJmol)

24 Molecular Parameters We also considered molecularparameters of analyzed compounds which were calculatedwith Dragon and Gaussian software Molecular descriptor isthe final result of a logic and mathematical procedure whichtransforms chemical information encoded within a symbolicrepresentation of a molecule into a useful number or a resultof a standardized experiment [23] Those parameters were

(i) MLOGP Moriguchi octanol-water partition coeffi-cient

(ii) Balaban index (BI) Narumi topological index (NTI)total structure connectivity index (TSC) Wienerindex (WI) numerical parameters characterizingcompoundsrsquo topology

(iii) HOMO highest occupied molecular orbital(iv) LUMO lowest unoccupied molecular orbital(v) HOMO-LUMO gap (HL gap) the energy difference

between the HOMO and LUMO






















Table2Apartof

inform

ationsyste

m(10fro

m140ob

jects)

Num

ber119899119877

lgCM

CgC

MC119866119860119866

ads

MLO

GP

BINTI

WI

MW

HOMO

LUMO

HLgap

dip

ROG

TSC

MIC

[mML]

Class

12

1215

619

275

52202

0175

1397

12712

5275


1646

4908

028

16937

Weak

22

2223

601

271

54208

0711

1407

14099

5768


0103

5294

0266

3558

Weak

32

3238

598

269

56213

1216

1405

15485

6307


2314

5804

0254

3295

Weak

42

4241

574

265

58217

1697

1397

16871

6877


35474

6246

0243

1634

Weak

52

5249

555

261

60223

2157

1386

18257

7468


8628

6777

0234

0712

Weak

62

6258

534

257

62227

2599

1373

19644

8074

39266minus038544minus01864

1minus019903

12501

725

0226

0086

Medium

72

7265

512

253

64235

3025

1359

2103

8692


16201

7791

0218

0020

Goo

d8

28

272

489

249

66239

4349

1346

22416

9319


20472

8282

0211

0005

Goo

d9

29

281

475

245

68243

4748

1333

23803

9952


24499

8831

0205

0002

Goo

d10

210

292

453

241

70248

5136

132

2518

91059


29011

9333

0199

0017

Goo

d


Table3Decision

rules

Num

ber

Con

ditio

nattributes

Exam

ples

Streng

thCon

firmationmeasure


MC120574CM

CΓsdot10

6Asdot10

20Δ119866

ads119872

log119875

MW

WI

HLg

apBI

TSC

NTI

Decision

classgood

1le10

le0218

5403857

07500

2le11

ge3836

4903500

07005

3leminus017317

ge2103

4203000

08000

4leminus015191

le0199

4203000

07142

5le11

le0199

3902785

07058

6ge6

ge40

669

leminus017314

3702642

06315

7ge7

ge37863

ge12

4234

02428

06315

8ge7

le483

ge12

4231

02214

06315

9ge7

ge277le694

2902071

066

6610

[711]

ge262

2902071

066

66Decision

classweak

11le6

ge248

2601857

09523

12le6

ge252

2501785

09523

13le6

le2814

2401714

09523

14le6

ge253

2401714

09523

15le6

le254

2301642

09523

16le6

ge256

2201571

09523

17le5

ge501

2201571

09523

18le5

ge252

2101500

09545

19le5

ge521

2001428

09523

20le5

le232

2001428

09523












119894⊳ 119864


119894⊳ 119864) which in this


119894⊳






000002004006

logC

MC

Gad

sRG

yrat

ion

HO

MO

LUM

Odi

pole

Bala

ban

Wie

ner

gCM

CM

WTS

CN

arum

iH

OM

O-L

UM

O g

ap

MLO

GP

minus002

minus004

minus006

A n G R

Con

firm

atio

n m

easu

reS


000

005

logC

MC

Gad

s

RGyr

atio

n

HO

MO

LUM

O

dipo

le

Bala

ban

Wie

ner

gCM

C

MW

TSC

Nar

umi

HO

MO

-LU

MO

gap

MLO

GP

minus005

minus010

A GR n

Con

firm

atio

n m

easu

reS






4 Conclusions




Acknowledgment


References












































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology






















Table2Apartof

inform

ationsyste

m(10fro

m140ob

jects)

Num

ber119899119877

lgCM

CgC

MC119866119860119866

ads

MLO

GP

BINTI

WI

MW

HOMO

LUMO

HLgap

dip

ROG

TSC

MIC

[mML]

Class

12

1215

619

275

52202

0175

1397

12712

5275


1646

4908

028

16937

Weak

22

2223

601

271

54208

0711

1407

14099

5768


0103

5294

0266

3558

Weak

32

3238

598

269

56213

1216

1405

15485

6307


2314

5804

0254

3295

Weak

42

4241

574

265

58217

1697

1397

16871

6877


35474

6246

0243

1634

Weak

52

5249

555

261

60223

2157

1386

18257

7468


8628

6777

0234

0712

Weak

62

6258

534

257

62227

2599

1373

19644

8074

39266minus038544minus01864

1minus019903

12501

725

0226

0086

Medium

72

7265

512

253

64235

3025

1359

2103

8692


16201

7791

0218

0020

Goo

d8

28

272

489

249

66239

4349

1346

22416

9319


20472

8282

0211

0005

Goo

d9

29

281

475

245

68243

4748

1333

23803

9952


24499

8831

0205

0002

Goo

d10

210

292

453

241

70248

5136

132

2518

91059


29011

9333

0199

0017

Goo

d


Table3Decision

rules

Num

ber

Con

ditio

nattributes

Exam

ples

Streng

thCon

firmationmeasure


MC120574CM

CΓsdot10

6Asdot10

20Δ119866

ads119872

log119875

MW

WI

HLg

apBI

TSC

NTI

Decision

classgood

1le10

le0218

5403857

07500

2le11

ge3836

4903500

07005

3leminus017317

ge2103

4203000

08000

4leminus015191

le0199

4203000

07142

5le11

le0199

3902785

07058

6ge6

ge40

669

leminus017314

3702642

06315

7ge7

ge37863

ge12

4234

02428

06315

8ge7

le483

ge12

4231

02214

06315

9ge7

ge277le694

2902071

066

6610

[711]

ge262

2902071

066

66Decision

classweak

11le6

ge248

2601857

09523

12le6

ge252

2501785

09523

13le6

le2814

2401714

09523

14le6

ge253

2401714

09523

15le6

le254

2301642

09523

16le6

ge256

2201571

09523

17le5

ge501

2201571

09523

18le5

ge252

2101500

09545

19le5

ge521

2001428

09523

20le5

le232

2001428

09523












119894⊳ 119864


119894⊳ 119864) which in this


119894⊳






000002004006

logC

MC

Gad

sRG

yrat

ion

HO

MO

LUM

Odi

pole

Bala

ban

Wie

ner

gCM

CM

WTS

CN

arum

iH

OM

O-L

UM

O g

ap

MLO

GP

minus002

minus004

minus006

A n G R

Con

firm

atio

n m

easu

reS


000

005

logC

MC

Gad

s

RGyr

atio

n

HO

MO

LUM

O

dipo

le

Bala

ban

Wie

ner

gCM

C

MW

TSC

Nar

umi

HO

MO

-LU

MO

gap

MLO

GP

minus005

minus010

A GR n

Con

firm

atio

n m

easu

reS






4 Conclusions




Acknowledgment


References












































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


Table2Apartof

inform

ationsyste

m(10fro

m140ob

jects)

Num

ber119899119877

lgCM

CgC

MC119866119860119866

ads

MLO

GP

BINTI

WI

MW

HOMO

LUMO

HLgap

dip

ROG

TSC

MIC

[mML]

Class

12

1215

619

275

52202

0175

1397

12712

5275


1646

4908

028

16937

Weak

22

2223

601

271

54208

0711

1407

14099

5768


0103

5294

0266

3558

Weak

32

3238

598

269

56213

1216

1405

15485

6307


2314

5804

0254

3295

Weak

42

4241

574

265

58217

1697

1397

16871

6877


35474

6246

0243

1634

Weak

52

5249

555

261

60223

2157

1386

18257

7468


8628

6777

0234

0712

Weak

62

6258

534

257

62227

2599

1373

19644

8074

39266minus038544minus01864

1minus019903

12501

725

0226

0086

Medium

72

7265

512

253

64235

3025

1359

2103

8692


16201

7791

0218

0020

Goo

d8

28

272

489

249

66239

4349

1346

22416

9319


20472

8282

0211

0005

Goo

d9

29

281

475

245

68243

4748

1333

23803

9952


24499

8831

0205

0002

Goo

d10

210

292

453

241

70248

5136

132

2518

91059


29011

9333

0199

0017

Goo

d


Table3Decision

rules

Num

ber

Con

ditio

nattributes

Exam

ples

Streng

thCon

firmationmeasure


MC120574CM

CΓsdot10

6Asdot10

20Δ119866

ads119872

log119875

MW

WI

HLg

apBI

TSC

NTI

Decision

classgood

1le10

le0218

5403857

07500

2le11

ge3836

4903500

07005

3leminus017317

ge2103

4203000

08000

4leminus015191

le0199

4203000

07142

5le11

le0199

3902785

07058

6ge6

ge40

669

leminus017314

3702642

06315

7ge7

ge37863

ge12

4234

02428

06315

8ge7

le483

ge12

4231

02214

06315

9ge7

ge277le694

2902071

066

6610

[711]

ge262

2902071

066

66Decision

classweak

11le6

ge248

2601857

09523

12le6

ge252

2501785

09523

13le6

le2814

2401714

09523

14le6

ge253

2401714

09523

15le6

le254

2301642

09523

16le6

ge256

2201571

09523

17le5

ge501

2201571

09523

18le5

ge252

2101500

09545

19le5

ge521

2001428

09523

20le5

le232

2001428

09523












119894⊳ 119864


119894⊳ 119864) which in this


119894⊳






000002004006

logC

MC

Gad

sRG

yrat

ion

HO

MO

LUM

Odi

pole

Bala

ban

Wie

ner

gCM

CM

WTS

CN

arum

iH

OM

O-L

UM

O g

ap

MLO

GP

minus002

minus004

minus006

A n G R

Con

firm

atio

n m

easu

reS


000

005

logC

MC

Gad

s

RGyr

atio

n

HO

MO

LUM

O

dipo

le

Bala

ban

Wie

ner

gCM

C

MW

TSC

Nar

umi

HO

MO

-LU

MO

gap

MLO

GP

minus005

minus010

A GR n

Con

firm

atio

n m

easu

reS






4 Conclusions




Acknowledgment


References












































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


Table3Decision

rules

Num

ber

Con

ditio

nattributes

Exam

ples

Streng

thCon

firmationmeasure


MC120574CM

CΓsdot10

6Asdot10

20Δ119866

ads119872

log119875

MW

WI

HLg

apBI

TSC

NTI

Decision

classgood

1le10

le0218

5403857

07500

2le11

ge3836

4903500

07005

3leminus017317

ge2103

4203000

08000

4leminus015191

le0199

4203000

07142

5le11

le0199

3902785

07058

6ge6

ge40

669

leminus017314

3702642

06315

7ge7

ge37863

ge12

4234

02428

06315

8ge7

le483

ge12

4231

02214

06315

9ge7

ge277le694

2902071

066

6610

[711]

ge262

2902071

066

66Decision

classweak

11le6

ge248

2601857

09523

12le6

ge252

2501785

09523

13le6

le2814

2401714

09523

14le6

ge253

2401714

09523

15le6

le254

2301642

09523

16le6

ge256

2201571

09523

17le5

ge501

2201571

09523

18le5

ge252

2101500

09545

19le5

ge521

2001428

09523

20le5

le232

2001428

09523












119894⊳ 119864


119894⊳ 119864) which in this


119894⊳






000002004006

logC

MC

Gad

sRG

yrat

ion

HO

MO

LUM

Odi

pole

Bala

ban

Wie

ner

gCM

CM

WTS

CN

arum

iH

OM

O-L

UM

O g

ap

MLO

GP

minus002

minus004

minus006

A n G R

Con

firm

atio

n m

easu

reS


000

005

logC

MC

Gad

s

RGyr

atio

n

HO

MO

LUM

O

dipo

le

Bala

ban

Wie

ner

gCM

C

MW

TSC

Nar

umi

HO

MO

-LU

MO

gap

MLO

GP

minus005

minus010

A GR n

Con

firm

atio

n m

easu

reS






4 Conclusions




Acknowledgment


References












































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology












119894⊳ 119864


119894⊳ 119864) which in this


119894⊳






000002004006

logC

MC

Gad

sRG

yrat

ion

HO

MO

LUM

Odi

pole

Bala

ban

Wie

ner

gCM

CM

WTS

CN

arum

iH

OM

O-L

UM

O g

ap

MLO

GP

minus002

minus004

minus006

A n G R

Con

firm

atio

n m

easu

reS


000

005

logC

MC

Gad

s

RGyr

atio

n

HO

MO

LUM

O

dipo

le

Bala

ban

Wie

ner

gCM

C

MW

TSC

Nar

umi

HO

MO

-LU

MO

gap

MLO

GP

minus005

minus010

A GR n

Con

firm

atio

n m

easu

reS






4 Conclusions




Acknowledgment


References












































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology


000002004006

logC

MC

Gad

sRG

yrat

ion

HO

MO

LUM

Odi

pole

Bala

ban

Wie

ner

gCM

CM

WTS

CN

arum

iH

OM

O-L

UM

O g

ap

MLO

GP

minus002

minus004

minus006

A n G R

Con

firm

atio

n m

easu

reS


000

005

logC

MC

Gad

s

RGyr

atio

n

HO

MO

LUM

O

dipo

le

Bala

ban

Wie

ner

gCM

C

MW

TSC

Nar

umi

HO

MO

-LU

MO

gap

MLO

GP

minus005

minus010

A GR n

Con

firm

atio

n m

easu

reS






4 Conclusions




Acknowledgment


References












































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology






































Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology








Volume 2014

Zoology






















Advances in

Virolog y



Volume 2014




Enzyme Research



Microbiology

Documents

Prediction of Antifungal Activity of Gemini Imidazolium Compounds