Construction of an optimal immune network model based on the modified swarm algorithm

  • G. A. Samigulina Kazakh-British technical university, Institute of information and computational technologies CS MES RK
  • Zh. A. Massimkanova Kazakh-British technical university, Institute of information and computational technologies CS MES RK


the approach of artificial immune systems, for the computer molecular design of new drugs andprediction the "structure-property/activity"relationship (QSAR) of chemical compounds is anactual problem. The article is devoted to the solution of the task of QSAR on the construction ofimmune network model based on the choice of optimal set of descriptors to facilitate the selection ofnew chemical compounds for candidate drugs with predefined properties. According to the conceptof multialgorithmic approach development of optimal immune network model and allocation ofinformative descriptors is carried out on the basis of swarm intelligence algorithms. In this workcomparison of standard particle swarm optimization algorithm (PSO) and modified inertia weightparticle swarm optimization (IWPSO) is described for selection of informative descriptors on theexample of drug compounds of the sulfanilamide group with various pharmacological activities. Thechoice of the parameters (fitness functions, population size, the number of iterations, etc.), whichdefine performance of the offered algorithms for creation of optimal set of descriptors is analysed.The results of modeling of dependence of fitness function values on the number of iterations insoftware products WEKA and Yarpiz (PSO) are given.


[1] Ahmad I., «Feature Selection Using Particle Swarm Optimization in Intrusion Detection», International Journal of Distributed Sensor Networks 11 (2015), accessed on March 28,2018,
[2] Bin J., Zhigang L. and Xingsheng G., «A dynamic inertia weight particle swarm optimization algorithm», Chaos, Solitons and Fractals 37 (2008): 698-705.
[3] Carkli Y.B.,Sertkaya C. and Yurtay N., «Prediction of secondary structures of hemoglobin using clonal selection algorithm», Proc. 7th International Workshop on Computer Science and Engineering(2017): 1387-1391.
[4] Farsangi M.M., Nezamabadi-pour H. and Lee K.Y., «Implementation of GCPSO for Multi-objective VAr Planning with SVC and Its Comparison with GA and PSO», Intelligent Systems Applications to Power Systems (2007), accessed March 15, 2018, DOI:10.1109/ISAP.2007.4441632
[5] Goodarzi M. and Dejaegher B., «Feature selection methods in QSAR studies», Journal AOAC Int 95(3) (2012): 636-651.
[6] Guo L. et al., «A comparison of various optimization algorithms of protein-ligand docking programs by fitness accuracy», Journal Molecular modeling 20(7) (2014): 2251.
[7] Hall M.A., «Correlation-based Feature Selection for Machine Learning». PhD diss., The University of Waikato, 1999.
[8] Jena P.K. and Parhi D.R., «A Modified Particle Swarm Optimization Technique for Crack Detection in Cantilever
Beams», Arabian Journal for Science and Engineering 40 (2015): 3263–3272.
[9] Jia Y. et al., «Generating Software Test Data by Particle Swarm Optimization», Materials of Asia-Pacific Conference on Simulated Evolution and Learning (2014): 37-47.
[10] Jukasik S. and Kowalski P., «Fully Informed Swarm Optimization Algorithms: Basic Concepts, Variants and Experimental Evaluation», Proceedings of the 2014 Federated Conference on Computer Science and Information Systems 2 (2014):155–161.
[11] Kennedy J. and Eberhart R.C., «Particle swarm optimization», IEEE International Conference on Neural Network (1995): 1942-1948.
[12] Khan S.U. et al., «A Modified Particle Swarm Optimization Algorithm for Global Optimizations of Inverse Problems», IEEE Transactions on Magnetics 52 (2016), accessed March 17, 2018, DOI:10.1109/TMAG.2015.2487678.
[13] Lasisi A., Ghazali R. and Chiroma H., «Utilizing clonal selection theory inspired algorithms and K-means clustering for predicting OPEC carbon dioxide emissions from petroleum consumption», Advances in Intelligent Systems and Computing 549 (2017): 101-110.
[14] Liefooghe A. et al., «ParadisEO-MOEO: A Software Framework for Evolutionary Multi-objective Optimization», Studies in Computational Intelligence 272 (2010): 87-117.
[15] Liu F. and Zhou Z., «A new data classification method based on chaotic particle swarm optimization and least squaresupport vector machine», Сhemometrics and intelligent laboratory systems (2015): 147-156.
[16] Moraglio A., Di Chio C. and Poli R., «Geometric Particle Swarm Optimization», EuroGP, LNCS 445 (2007): 125-135.
[17] Mu A., Cao D. and Wang X., «A Modified Particle Swarm Optimization Algorithm», Natural Science 1 (2009):151-155.
[18] Pichuzhkina А.V., «Metod roya chastic v zadachah optimalnoi orientacii sputnikov». [Particle swarm optimization in problems of optimal orientation of satellites] Master diss., Moscow Institute of Physics and Technology, 2017.
[19] Samigulina G.A., Immune network modeling technology for complex objects intellectual control and forecasting system. USA: Science Book Publishing House, 2015, 172.
[20] Samigulina G.A. and Massimkanova Zh.A., «Computer modeling og new drugs based on the methods of swarm intelligence and immune network modeling», Bulletin of national technical university "KHPI" 50(1271) (2017): 88-94.
[21] Samigulina G.A. and Massimkanova Zh.A., «Ontologicheskie modeli algoritmov roevogo intellekta dlya immunnosetevogo modelirovaniya lekarstvennyh preparatov». [The ontological models of swarm intelligence algorithm for immunne network modeling of medical drugs] Bulletin of Al-Farabi Kazakh national university: Mathematics, Mechanics and Computer Science Series 1(93)(2017): 92-104.
[22] Samigulina G.A., Samigulina Z.I., «The construction of an optimal immune network model for forecasting the properties of unknown drug compounds on the basis of multialgorithmic approach», Problems of informatics 2 (2013): 21-29.
[23] Samigulina G.A., Samigulina Z.I., «Primenenie sovremennyh metodov Data Mining dlya prognozirovaniya zavisimosti "structura-svoistvo"khimicheskih soedineniyi sulfanilamodov», [Application of modern methods of Data Mining for prediction "structure/property"relationship of chemical compounds of sulfonamides] Problems of evolution of open systems 19 (2017): 99-107.
[24] Shen H., Zhao Y., «An improved cooperative PSO algorithm», Mechatronic Science, Electric Engineering and Computer (2011), accessed November 19, 2017, DOI: 10.1109/MEC.2011.6025643
[25] Shi Y. and Eberhart R.C., «A modified particle swarm optimizer», Proceedings of Congress on Evolutionary Computation (1998): 79-73.
[26] Thamaraichelvi B. and Yamuna G., «Hybrid firefly swarm intelligence based feature selection for medical data classification and segmentation in SVD - NSCT domain», International Journal of Advanced Research 4(9) (2016): 744-760.
[27] Timmis J., Neal M. and Hunt J., «An artificial immune system for data analysis», BioSystems 55 (1) (2000): 43-150.
[28] Umapathy P., Venkataseshaiah C. and Arumugam M.S., «Particle Swarm Optimization with Various Inertia Weight Variants for Optimal Power Flow Solution», Discrete Dynamics in Nature and Society (2010), accessed May 15, 2018,
[29] Zhan Z. et al., «Adaptive Particle Swarm Optimization», IEEE Transactions on Systems, Man, and Cybernetics 39 (2009): 1362-1381.
[30] «Molecular descriptors», accessed March 15, 2018,
[31] «Python implementation of Particle Swarm Optimization», accessed March 15, 2018,
[32] «Particle swarm optimization for non-linear programming», accessed November 19, 2017,
[33] «Tribes particle swarm optimization technique», accessed March 14, 2017,
How to Cite
SAMIGULINA, G. A.; MASSIMKANOVA, Zh. A.. Construction of an optimal immune network model based on the modified swarm algorithm. Journal of Mathematics, Mechanics and Computer Science, [S.l.], v. 98, n. 2, p. 77-87, aug. 2018. ISSN 1563-0277. Available at: <>. Date accessed: 20 feb. 2019. doi:
Keywords optimal immune network model, selection of informative descriptors, particle swarm optimization(PSO)