Xác định các chất điều chỉnh chính trong ung thư tuyến tụy bằng cách sử dụng thuật toán tiến hóa

UNIVERSITE NATIONALE DU VIETNAM, HANOI INSTITUT DE LA FRANCOPHONIE POUR L’INNOVATION (Renommé de l’Institut Francophone International, IFI) FOTSING SIKADIE GERVAIS SERTILANGE Identification des principaux régulateurs dans le cancer du pancréas en utilisant les algorithmes évolutifs Xác định chất điểu chỉnh ung thư tuyến tụy cách sử dụng thuật tốn tiến hóa MEMOIRE DE FIN D’ETUDES DU MASTER INFORMATIQUE HANOI – 2018 UNIVERSITE NATIONALE DU VIETNAM, HANOI INSTITUT DE LA FRANCOPHONIE POUR L’INNOVATION (Renommé de l’Institut Francophone International, IFI) FOTSING SIKADIE GERVAIS SERTILANGE Identification des principaux régulateurs dans le cancer du pancréas en utilisant les algorithmes évolutifs Xác định chất điểu chỉnh ung thư tuyến tụy cách sử dụng thuật tốn tiến hóa Spécialité: Systèmes intelligents et multimédia Code: Programme pilote MEMOIRE DE FIN D’ETUDES DU MASTER INFORMATIQUE- Sous la direction de: DR Claude PASQUIER et Dr Denis PALLEZ HANOI – 2018 ATTESTATION SUR L’HONNEUR J’atteste sur l’honneur que ce mémoire a été réalisé par moi-même et que les données et les résultats qui y sont présentés sont exacts et n’ont jamais été publiés ailleurs La source des informations citées dans ce mémoire a été bien précisée LỜI CAM ĐOAN Tôi cam đoan cơng trình nghiên cứu riêng tơi Các số liệu, kết nêu Luận văn trung th ực ch ưa t ừng đ ược cơng bố cơng trình khác Các thơng tin trích d ẫn Lu ận văn rõ nguồn gốc Signature de l’étudiant FOTSING SIKADIE GERVAIS Résumé Dans ce travail, on propose un pipeline qui permet de sélectionner les sous-ensembles de gènes liés au cancer du pancréas partant d’un jeux de données contenant des échantillons normaux et cancéreux Chaque échantillon est caractérisé par le niveau d’expression de chaque gènes (environ 24000 gènes) Pour réaliser cela, on considère critères que doivent satisfaire les sous-ensembles de gènes sélectionnés Le premier critère est le rôle du sous-ensemble sélectionné dans le cancer du pancréas Pour matérialiser mathématiquement cela, on utilise SVM pour construire un modèle de classification afin de calculer la précision du modèle considéré juste en considérant le sous ensemble de gènes sélectionnés comme variables explicatives Le second critère est que la taille du sous ensemble de gène retenu doit être aussi petit que possible ce qui permettra de d’alléger la tâche de vérification physique par les biologistes qui est par d’ailleurs très coûteuse Le dernier critère est la signification biologique des sous ensembles de gènes Pour matérialiser cela, les sous ensembles de gènes avec plusieurs interactions biologiques entre eux sont privilégiés On a utilisé un graphe non orienté d’interactions entre gènes(bien connu par les biologistes) donné par la plate-forme BioGRID Partant d’un sous ensemble de gènes, on calcule le sous-graphe d’interaction entre les gènes sélectionnés l’aide du graphe de d’interaction connu Ce problème est un problème de sélection de gènes mais peut être résumé plus généralement en un problème de feature selection Les données utilisées dans ce type de problème ont la particularité d’être de grande dimension( plus de 1000 gènes) et avec peux d’instances(moins de 100 individus) Dans ce travail, on présente le feature selection, puis on se concentre sur le feature selection en grande dimension et peux d’instances Nous sommes entré en détail dans types d’algorithmes intéressants proposés par la littérature (ISVM-RCE,LTS,NSGA2) Nous avons implémenté et testé ces algorithmes sur quelques jeux de données On a proposé quelques améliorations pour l’algorithme LTS Après on a proposé un pipeline pour la sélection des gènes importants dans le cancer du pancréas en utilisant l’algorithme génétique NSGA2 et LTS amélioré qu’on a proposé et baptisộ : MO-LTS MO-LTS est aussi testộ en remplaỗant NSGA2 dans le pipeline Un BioGRID est un référentiel d’interaction entre gènes avec des données compilées dans des fichiers texte sous forme de tableaux i travail de conformité biologique est réalisé pour les résultats fournis par les deux algorithmes, et la combinaison de NSGA2 et MO-LTS améliore grandement le temps de convergence vers la solution ii Abstract In this work, a pipeline is proposed to select subsets of pancreatic cancer genes from a dataset containing normal and cancerous samples Each sample is characterized by the level of expression of each gene (approximately 24000 genes) To achieve this, we consider criteria that the subsets of selected genes must satisfy The first criterion is the role of the selected subset in pancreatic cancer To mathematically materialize this, we use SVM to construct a classification model in order to calculate the precision of the considered model just by considering the subset of selected genes as explanatory variables The second criterion is that the size of the gene sub-set retained should be as small as possible, which will make it possible to lighten the task of physical verification by biologists, which is, by the way, very expensive The final criterion is the biological significance of subsets of genes To materialize this, the subsets of genes with several biological interactions between them are privileged An undirected graph of gene interactions (well known by biologists) was used by the BioGRID platform BioGRID is a repository of interaction between genes with data compiled in text files in tabular form Starting from a subset of genes, the subgraph of interaction between the selected genes is calculated using the known interaction graph This problem is a problem of gene selection but can be summarized more generally in a problem of feature selection The data used in this type of problem have the particularity of being large (more than 1000 genes) and with few instances (less than 100 individuals) In this work, we present the feature selection, then we focus on feature selection in large size and can instances We went into detail in types of interesting algorithms proposed by the literature (ISVM-RCE, LTS, NSGA2) We have implemented and tested these algorithms on some datasets Some improvements have been proposed for the LTS algorithm Then a pipeline was proposed for the selection of important genes in pancreatic cancer using the improved NSGA2 and LTS genetic algorithm proposed and named : MO-LTS MO-LTS is also tested by replacing NSGA2 in the pipeline A biological compliance work is done for the results provided by the two algorithms, and the combination of NSGA2 and MO-LTS greatly improves the convergence time to the solution iii Table des matières Résumé i Abstract iii Table des matières v Liste des tableaux vii Liste des figures ix Remerciements xi Introduction Étude bibliographique 1.1 Feature selection 1.2 Architecture des algorithmes de feature selection 1.3 Travaux connexes 1.4 Description de quelques algorithmes intéressants de feature selection 5 10 18 Comparaison des algorithmes existants 25 Améliorations proposées pour LTS 3.1 Stratégies d’initialisation 3.2 Méthode proposée : Multi Objectives-LTS (MO-LTS) 29 29 29 Application au jeu de données sur le cancer du pancréas (GSE) 4.1 Pipeline de feature selection grande échelle 4.2 Résultats avec les objectifs NSGA2, LTS et ISVM-RCE 4.3 Méthode proposée : NSGA2 pour la recherche globale et MO-LTS recherche locale 4.4 Expériences sur le côlon et GSE 4.5 Résultats 35 36 36 Conclusion pour la 40 41 43 47 v Bibliographie vi 49 Liste des tableaux 1.1 1.2 Algorithmes intéressants de feature selection adaptés aux jeux de données de grandes dimensions et peux d’instances paramètres de NSGA2 17 22 2.1 2.2 2.3 Description des jeux de données utilisés Paramètres de ISVM-RCE Paramètres de Learning Tabu Search 25 26 26 3.1 Comparison on 20 runs 34 4.1 genes obtained with our genes selection pipeline using Memetic and NSGA2 45 vii Conclusion Les résultats sont prometteurs lorsqu’on considère la densité Memetic apporte plus de vitesse l’algorithme de sélection de caractéristiques comparé NSGA2 Cela nous permet pour obtenir des résultats plus précis et plus rapides Memetic nous apporte la réduction de l’évaluation du nombre nécessaire et des solutions de qualité car elle combine recherche globale et locale Nous avons également mis en évidence des gènes déjà identifiés comme ayant un rôle dans le cancer du pancréas De nouveaux gènes détectés ont également été mis en évidence Les références aux gènes déjà identifiés dans la littérature ont été mentionnées Cela confirme le travail effectué dans ces articles qui ont utilisé d’autres approches NSGA2 et Memetic gènes identifiés qui n’ont jamais été ou ne sont pas souvent mentionnés comme liés PDAC ces gènes sont également répertoriés Le pipeline proposé a encore besoin de temps pour s’exécuter (semaines de calcul) Un mauvais point est que nous appliquons un filtre pour enlever les gènes qui n’ont pas d’impact sur la variable de sortie Les objectifs du stage ont clairement été atteints en l’occurrence : – Proposition d’une amélioration pour LTS en le rendant multi-objectif – Proposition d’un pipeline de sélection de gènes importants dans le cancer – Application de ce pipeline un jeux de données sur le cancer du pancréas – Proposition d’un algorithme memetique permettant d’améliorer la vitesse de recherche de solution dans l’identification des gènes importants – Vérification des résultats obtenus Mais ces gènes peuvent aussi être liés au cancer Donc, l’avenir, nous pouvons travailler sur : – Amélioration de l’algorithme pourqu’il puisse travailler sans filtre préalable – Identifier un objectif biologique plus significatif que la densité du graphe – Intégrer d’autres objectifs comme l’implication de gènes sélectionnés dans le processus biologique étudié 47 Bibliographie [1] The contribution of cohesin-sa1 to gene expression and chromatin architecture in two murine tissues 2015 [2] R Agrawal and R Bala A hybrid approach for selection of relevant features for microarray datasets International Journal of Computer and Information Engineering, 1(2), 2007 [3] D B Azizoglu, C Braitsch, D K Marciano, and O Cleaver Afadin and rhoa control pancreatic endocrine mass via lumen morphogenesis Genes & development, 31(23-24) :2376– 2390, 2017 [4] M C T C H K J T.-B M N B C P K J Y M B G M P H G T M Bagnati M, Ogunkolade BW Glucolipotoxicity initiates pancreatic -cell death through tnfr5/cd40mediated stat1 and nf-b activation 2016 [5] A G P S B J B M.-G J B A Barbosa HC, Bordin S Islet neogenesis-associated protein signaling in neonatal pancreatic rat islets : involvement of the cholinergic pathway 2008 [6] C A Barnes, E Krzywda, S Lahiff, D McDowell, K K Christians, P Knechtges, P Tolat, M Hohenwalter, K Dua, A H Khan, et al Development of a high risk pancreatic screening clinic using 3.0 t mri Familial cancer, 17(1) :101–111, 2018 [7] M T Barrett, R Deiotte, E Lenkiewicz, S Malasi, T Holley, L Evers, R G Posner, T Jones, H Han, M Sausen, et al Clinical study of genomic drivers in pancreatic ductal adenocarcinoma British Journal of Cancer, 117(4) :572, 2017 [8] O Basturk, M F Berger, H Yamaguchi, V Adsay, G Askan, U K Bhanot, A Zehir, F Carneiro, S.-M Hong, G Zamboni, et al Pancreatic intraductal tubulopapillary neoplasm is genetically distinct from intraductal papillary mucinous neoplasm and ductal adenocarcinoma Modern Pathology, 30(12) :1760, 2017 49 [9] L M G A S T L G.-P S A G Beil M, Leser J Caspase 8-mediated cleavage of plectin precedes f-actin breakdown in acinar cells during pancreatitis 2002 [10] W S L L Bian Y, Yu Y Up-regulation of fatty acid synthase induced by egfr/erk activation promotes tumor growth in pancreatic cancer 2015 [11] M E F P L A M G.-V F G A T O D J I J G M B D C M Birnbaum DJ, Adélaïde J Genome profiling of pancreatic adenocarcinoma 2011 [12] E Bonilla-Huerta, A Hernández-Montiel, R Morales-Caporal, and M Arjona-López Hybrid framework using multiple-filters and an embedded approach for an efficient selection and classification of microarray data IEEE/ACM transactions on computational biology and bioinformatics, 13(1) :12–26, 2016 [13] K R H H W C G Z.-B M C D V H D Borazanci E, Millis SZ Adenosquamous carcinoma of the pancreas : Molecular characterization of 23 patients along with a literature review 2015 [14] K N W K Brafman DA, Phung C Regulation of endodermal differentiation of human embryonic stem cells through integrin-ecm interactions 2013 [15] J B L Y H P A J.-L C W H Chang DZ, Ma Y Increased cdc20 expression is associated with pancreatic ductal adenocarcinoma differentiation and progression 2012 [16] L S W C W F Chang WS, Chang NT Tissue-specific cancer-related serpin gene cluster at human chromosome band 3q26 2000 [17] J Chen, S Zaidi, S Rao, J.-S Chen, L Phan, P Farci, X Su, K Shetty, J White, F Zamboni, et al Analysis of genomes and transcriptomes of hepatocellular carcinomas identifies mutations and gene expression changes in the transforming growth factor-β pathway Gastroenterology, 154(1) :195–210, 2018 [18] J K D K P B H J.-E H H M W J T J Y A Chen KT, Kim PD Potential prognostic biomarkers of pancreatic cancer 2014 [19] I Dando, R Pacchiana, E Dalla Pozza, I Cataldo, S Bruno, P Conti, M Cordani, A Grimaldi, G Butera, M Caraglia, et al Ucp2 inhibition induces ros/akt/mtor axis : Role of gapdh nuclear translocation in genipin/everolimus anticancer synergism Free Radical Biology and Medicine, 113 :176–189, 2017 50 [20] K Deb, A Pratap, S Agarwal, and T Meyarivan A fast and elitist multiobjective genetic algorithm : Nsga-ii IEEE transactions on evolutionary computation, 6(2) :182–197, 2002 [21] C Y P L B M S J Dong Z, Shang H Sulforaphane protects pancreatic acinar cell injury by modulating nrf2-mediated oxidative stress and nlrp3 inflammatory pathway 2016 [22] T T S A F G S F M Y Farkas G Jr1, Tiszlavicz Z Analysis of plasma levels and polymorphisms of s100a8/9 and s100a12 in patients with acute pancreatitis 2014 [23] M D Ferguson, L Dong, J Wan, J L Deneve, P V Dickson, S W Behrman, D Shibata, M G Martin, and E S Glazer Molecular alterations associated with dna repair in pancreatic adenocarcinoma are associated with sites of recurrence Journal of Gastrointestinal Cancer, pages 1–7, 2018 [24] D Frodyma, B Neilsen, D Costanzo-Garvey, K Fisher, and R Lewis Coordinating erk signaling via the molecular scaffold kinase suppressor of ras F1000Research, 6, 2017 [25] T E Y S H T Y M S N S K K N S K Furukawa T, Kuboki Y Whole-exome sequencing uncovers frequent gnas mutations in intraductal papillary mucinous neoplasms of the pancreas Gastroenterology, 2011 [26] I A Gheyas and L S Smith Feature subset selection in large dimensionality domains Pattern recognition, 43(1) :5–13, 2010 [27] R M H N S P A M M S P J M A G M K S K E M P S T H M M D J W K G Giessrigl B, Krieger S Hsp90 stabilizes cdc25a and counteracts heat shock-mediated cdc25a degradation and cell-cycle attenuation in pancreatic carcinoma cells 2008 [28] K B V J M D T Z H M G R V G C J L K L M Gleeson FC, Kerr SE Targeted next generation sequencing of endoscopic ultrasound acquired cytology from ampullary and pancreatic adenocarcinoma has the potential to aid patient stratification for optimal therapy selection 2016 [29] J S W D F K Y D Z X X L W K L F Gong J, Dong H Fenugreek lactone attenuates palmitate-induced apoptosis and dysfunction in pancreatic -cells World Journal of gastroentorology, 2015 [30] B A D N H D K J A M Z G G S Grant RC, Holter S Comparison of practice guidelines, brcapro, and genetic counselor estimates to identify germline brca1 and brca2 mutations in pancreatic cancer Journal of Genetic Counseling, 1(2), 2018 51 [31] Z.-D Guo, L Zhao, P Wang, W.-H Deng, Q Shi, T Zuo, Y.-P Hong, and W.-X Wang Fetal liver injury ameliorated by migration inhibitory factor inhibition in a rat model of acute pancreatitis in pregnancy Journal of Obstetrics and Gynaecology Research, 2017 [32] I Guyon, J Weston, S Barnhill, and V Vapnik Gene selection for cancer classification using support vector machines Machine learning, 46(1) :389–422, 2002 [33] K H C S L S P K J H K Y K Y Han D, Moon S Detection of differential proteomes associated with the development of type diabetes in the zucker rat model using the itraq technique J Proteome Res., 2014 [34] K M M H F K H K K M K Y K T T H H K I H Hasegawa K, Wakino S The hydrolase ddah2 enhances pancreatic insulin secretion by transcriptional regulation of secretagogin through a sirt1-dependent mechanism in mice 2013 [35] S D Hedman AC, Smith JM The biology of iqgap proteins : beyond the cytoskeleton Journal of Gastrointestinal Cancer, pages 1–7, 2015 [36] J M Herman, S K Jabbour, S H Lin, M P Deek, C C Hsu, E K Fishman, S Kim, J L Cameron, M Chekmareva, D A Laheru, et al Smad4 loss correlates with higher rates of local and distant failure in pancreatic adenocarcinoma patients receiving adjuvant chemoradiation Pancreas, 47(2) :208–212, 2018 [37] Y K Q Z M K A K S T Y K H M K K K S Higuchi Y, Shiraki N Synthesized basement membranes direct the differentiation of mouse embryonic stem cells into pancreatic lineages 2010 [38] G M O N L A L C H R G M Hong SM, Kelly D Multiple genes are hypermethylated in intraductal papillary mucinous neoplasms of the pancreas Current drug targets, 2008 [39] G G G K H J Z M K S W V W L S A S K S I W M Huang X, Dai FF The identification of novel proteins that interact with the glp-1 receptor and restrain its activity 2013 [40] W X Z S Huang YK, Liu H Annexin a2 and cd105 expression in pancreatic ductal adenocarcinoma is associated with tumor recurrence and prognosis 2014 [41] K H C H C T P Y Hwang KW, Won TJ Erratum to “characterization of the regulatory roles of the sumo 2012 52 [42] I C B R S P V U H E Iuga C, Seicean A Proteomic identification of potential prognostic biomarkers in resectable pancreatic ductal adenocarcinoma Current drug targets, 2014 [43] P A C G Jäggi F, Cabrita MA Modulation of endocrine pancreas development but not beta-cell carcinogenesis by sprouty4 2008 [44] K Kawahito, H Sato, M Kadosaki, A Egawa, and Y Misawa Spike in glucose levels after reperfusion during aortic surgery : assessment by continuous blood glucose monitoring using artificial endocrine pancreas General thoracic and cardiovascular surgery, pages 1–5, 2017 [45] V I V E A S A P A L A F B V B N B L B K B D B A C M C M C M C I D S D L E A F A F S F J G L G U G J G A G F G N G Z H A H M H J H S H J H H H X I T K K K M K D L S L P L E M Y M F M A M I M D M L M N M M M G M K M G O R P F P T P J P E R J R C R.-Q P R A S A S M S J S R S J S A T E T A T S V P V F W S Y H Z W Z L K G G L Kepp O, Senovilla L Consensus guidelines for the detection of immunogenic cell death 2014 [46] N Kubo, K Araki, B Altan, K Hoshino, N Ishii, M Tsukagoshi, T Igarashi, A Watanabe, T Kato, K Hirai, et al Enhanced karyopherin-α2 expression is associated with carcinogenesis in patients with intraductal papillary mucinous neoplasms Pancreatology, 17(4) :611–616, 2017 [47] R M R T U W T A S W M J Köhler CU, Kreuter A Validation of different replication markers for the detection of beta-cell proliferation in human pancreatic tissue 2010 [48] L L Ductal adenocarcinoma of the pancreas usually retained smad4 and p53 protein status as well as expression of epithelial-to-mesenchymal transition markers and cell cycle regulators at the stage of liver metastasis 2014 [49] B N B V L E C J R C J M N Z Y C S J W J R I Y T B J Lal S, Burkhart RA Hur posttranscriptionally regulates wee1 : implications for the dna damage response in pancreatic cancer cells 2016 [50] C S D E M.-L N P M Y C W J B J Lal S, Zarei M Wee1 inhibition in pancreatic cancer cells is dependent on dna repair status in a context dependent manner 2016 53 [51] W M C J R E W K d l M J G F P R P M P B V S C S Levy MJ, Topazian MD Initial evaluation of the efficacy and safety of endoscopic ultrasound-guided direct ganglia neurolysis and block 2014 [52] T J X F W J Z R Li J, Li Q Quantitative phosphoproteomics revealed glucosestimulated responses of islet associated with insulin secretion 2015 [53] M Q L J X Q H L D W L Y W F R K W E Li W, Wu Z Hyperglycemia regulates txnip/trx/ros axis via p38 mapk and erk pathways in pancreatic cancer 2014 [54] T.-C Lin, R.-S Liu, Y.-T Chao, and S.-Y Chen Classifying subtypes of acute lymphoblastic leukemia using silhouette statistics and genetic algorithms Gene, 518(1) :159–163, 2013 [55] Lin-Kai Luo, Deng-Feng Huang, Ling-Jun Ye, Qi-Feng Zhou, Gui-Fang Shao, and Hong Peng Improving the Computational Efficiency of Recursive Cluster Elimination for Gene Selection IEEE/ACM Transactions on Computational Biology and Bioinformatics, 8(1) :122–129, jan 2011 [56] W Liu, B Zhang, Q Hu, Y Qin, W Xu, S Shi, C Liang, Q Meng, J Xiang, D Liang, et al A new facet of ndrg1 in pancreatic ductal adenocarcinoma : suppression of glycolytic metabolism International journal of oncology, 50(5) :1792–1800, 2017 [57] Q Z Z C Liu Y, Liao R Pro-inflammatory cytokine-driven pi3k/akt/sp1 signalling and h2s production facilitates the pathogenesis of severe acute pancreatitis 2017 [58] K E D.-J K S O Lundberg M, Krogvold L Expression of interferon-stimulated genes in insulitic pancreatic islets of patients recently diagnosed with type diabetes 2016 [59] Y Luo, Z Yang, D Li, Z Liu, L Yang, Q Zou, and Y Yuan Ldhb and fabp4 are associated with progression and poor prognosis of pancreatic ductal adenocarcinomas Applied Immunohistochemistry & Molecular Morphology, 25(5) :351–357, 2017 [60] E Y U M H T K K T S M K S M S T F H N K N K Y M K T I T E H N H M E Maekawa T, Kamada Y Genome profiling of pancreatic adenocarcinoma 2016 [61] V E D M R T B V H F P S S E P A Marinoni I, Kurrer AS Ribosomal biogenesis genes play an essential and p53-independent role in zebrafish pancreas development 2014 54 [62] O L W T N A Markljung E, Adamovic T A rare microduplication in a familial case of annular pancreas and duodenal stenosis 2012 [63] F A L Marson, C S Bertuzzo, T K de Araujo, T D R Hortencio, A F Ribeiro, and J D Ribeiro Pancreatic insufficiency in cystic fibrosis : Influence of inflammatory response genes Pancreas, 47(1) :99–109, 2018 [64] F A L Marson, C S Bertuzzo, T K de Araujo, T D R Hortencio, A F Ribeiro, and J D Ribeiro Pancreatic insufficiency in cystic fibrosis : Influence of inflammatory response genes Pancreas, 47(1) :99–109, 2018 [65] S S Mello, L J Valente, N Raj, J A Seoane, B M Flowers, J McClendon, K T BiegingRolett, J Lee, D Ivanochko, M M Kozak, et al A p53 super-tumor suppressor reveals a tumor suppressive p53-ptpn14-yap axis in pancreatic cancer Cancer cell, 32(4) :460–473, 2017 [66] C Melzer, R Hass, J von der Ohe, H Lehnert, and H Ungefroren The role of tgf-β and its crosstalk with rac1/rac1b signaling in breast and pancreas carcinoma Cell Communication and Signaling, 15(1) :19, 2017 [67] K Michalova, M Michal, D V Kazakov, M Sedivcova, O Hes, L Hadravsky, A Agaimy, M Tretiakova, C Bacchi, A Hartmann, et al Primary signet ring stromal tumor of the testis : a study of 13 cases indicating their phenotypic and genotypic analogy to pancreatic solid pseudopapillary neoplasm Human pathology, 67 :85–93, 2017 [68] G C F S B C S C Miron A, Calu V Laparoscopically treated pancreatic insulinoma case report 2010 [69] J D J S K N P N Mishra PK, Raghuram GV Mitochondrial oxidative stress-induced epigenetic modifications in pancreatic epithelial cells 2014 [70] L Mousin, L Jourdan, and C Dhaenens Feature Selection using Tabu Search with Learning Memory : Learning Tabu Search pages 141–156, may 2016 [71] S P V P O.-L E B A R E Nagaraj V, King B Complement inhibitor cd55 governs the integrity of membrane rafts in pancreatic beta cells, but plays no role in insulin secretion 2015 [72] S J L N L H Z D L C D J Z Z C M H Q C C L C W C P Q C Y Y X S J Ning SL, Zheng WS Different downstream signalling of cck1 receptors regulates distinct functions of cck in pancreatic beta cells 2015 55 [73] K M K T Nishida Y, Aida K Antibody-validated proteins in inflamed islets of fulminant type diabetes profiled by laser-capture microdissection followed by mass spectrometry 2014 [74] K M K T Nishida Y1, Aida K2 Antibody-validated proteins in inflamed islets of fulminant type diabetes profiled by laser-capture microdissection followed by mass spectrometry 2014 [75] Y L Z J G S F Y M Y W J Peng YP, Zhu Y The expression and prognostic roles of mcms in pancreatic cancer 2016 [76] L S Provost E, Weier CA Multiple ribosomal proteins are expressed at high levels in developing zebrafish endoderm and are required for normal exocrine pancreas development 2013 [77] Z X A F N E G R P M L S Provost E, Wehner KA Ribosomal biogenesis genes play an essential and p53-independent role in zebrafish pancreas development 2012 [78] B Qu, L Yao, H.-l Ma, H.-l Chen, Z Zhang, and J Xie Prognostic significance of autophagy-related proteins expression in resected human gastric adenocarcinoma Journal of Huazhong University of Science and Technology [Medical Sciences], 37(1) :37–43, 2017 [79] D Rajamani and M K Bhasin Identification of key regulators of pancreatic cancer progression through multidimensional systems-level analysis Genome Medicine, 8(1) :38, 2016 [80] S Rao, R A Beckman, S Riazi, C S Yabar, S M Boca, J L Marshall, M J Pishvaian, J R Brody, and S Madhavan Quantification and expert evaluation of evidence for chemopredictive biomarkers to personalize cancer treatment Oncotarget, 8(23) :37923, 2017 [81] N S K A Rauhala HE, Teppo S Silencing of the arp2/3 complex disturbs pancreatic cancer cell migration 2013 [82] M L Raymer, W F Punch, E D Goodman, L A Kuhn, and A K Jain Dimensionality reduction using genetic algorithms IEEE transactions on evolutionary computation, 4(2) :164–171, 2000 [83] P G Rotti, W Xie, A Poudel, Y Yi, X Sun, S R Tyler, A Uc, A W Norris, M Hara, J F Engelhardt, et al Pancreatic and islet remodeling in cystic fibrosis transmembrane conductance regulator (cftr) knockout ferrets The American journal of pathology, 2018 56 [84] K H H M B M S J Ryschich E, Huszty G Transferrin receptor is a marker of malignant phenotype in human pancreatic cancer and in neuroendocrine carcinoma of the pancreas 2014 [85] L S Genetic predisposition and environmental risk factors to pancreatic cancer : A review of the literature 2009 [86] C L P N Sae-Lee C, Moolsuwan K Chrebp regulates itself and metabolic genes implicated in lipid accumulation in -cell line BioMed research international, 2016 [87] N Safari-Alighiarloo, M Taghizadeh, S M Tabatabaei, S Shahsavari, S Namaki, S Khodakarim, and M Rezaei-Tavirani Identification of new key genes for type diabetes through construction and analysis of protein–protein interaction networks based on blood and pancreatic islet transcriptomes Journal of diabetes, 9(8) :764–777, 2017 [88] S J Sajjadi, X Qian, B Zeng, and A A Adl Network-Based Methods to Identify Highly Discriminating Subsets of Biomarkers IEEE/ACM Transactions on Computational Biology and Bioinformatics, 11(6) :1029–1037, nov 2014 [89] N H N B M S Salter NW, Ande SR Functional characterization of naturally occurring transglutaminase mutants implicated in early-onset type diabetes 2012 [90] W J Sans MD Calcineurin is required for translational control of protein synthesis in rat pancreatic acini 2004 [91] D M Saulino, P S Younes, J M Biley, and M Younes Expression of exportin (xpo1/crm1) in pancreatic adenocarcinoma., 2017 [92] R S F E S N B H S R H M Schwer CI, Stoll P Carbon monoxide releasing molecule-2 corm-2 represses global protein synthesis by inhibition of eukaryotic elongation factor eef2 2013 [93] Y Shi, Z R Qian, S Zhang, W Li, Y Masugi, T Li, J A Chan, J Yang, A Da Silva, M Gu, et al Cell cycle protein expression in neuroendocrine tumors : Association of cdk4/cdk6, ccnd1, and phosphorylated retinoblastoma protein with proliferative index Pancreas, 46(10) :1347–1353, 2017 [94] J Shirakawa, M Fernandez, T Takatani, A El Ouaamari, P Jungtrakoon, E R Okawa, W Zhang, P Yi, A Doria, and R N Kulkarni Insulin signaling regulates the foxm1/plk1/cenp-a pathway to promote adaptive pancreatic β cell proliferation Cell metabolism, 25(4) :868–882, 2017 57 [95] R K Somvanshi, A Jhajj, M Heer, and U Kumar Characterization of somatostatin receptors and associated signaling pathways in pancreas of r6/2 transgenic mice Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease, 1864(2) :359–373, 2018 [96] T A G.-G E N A S.-G A L.-R F Z X S A T J R M F C J A P A H M Song D, Chaerkady R Antitumor activity and molecular effects of the novel heat shock protein 90 inhibitor, ipi-504, in pancreatic cancer 2008 [97] K H M T K K Song JY, Lim JW Oxidative stress induces nuclear loss of dna repair proteins ku70 and ku80 and apoptosis in pancreatic acinar ar42j cells The Journal of Biomedical chemistry., 2014 [98] D Tang, Q Wu, J Zhang, H Zhang, Z Yuan, J Xu, Y Chong, Y Huang, Q Xiong, S Wang, et al Galectin-1 expression in activated pancreatic satellite cells promotes fibrosis in chronic pancreatitis/pancreatic cancer via the tgf-β1/smad pathway Oncology reports, 2018 [99] Y Tang, Y.-Q Zhang, and Z Huang Development of two-stage svm-rfe gene selection strategy for microarray expression data analysis IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB), 4(3) :365–381, 2007 [100] C G B M M F S B T J.-B F Tapia-Limonchi R, Díaz I Impact of exposure to low concentrations of nitric oxide on protein profile in murine and human pancreatic islet cells 2014 [101] S Teoh and S Das Notch signalling pathways and their importance in the treatment of cancers Current drug targets, 2017 [102] J D A G S S S M S I M Thorpe H, Akhlaq M Multiple pathways regulate cten in colorectal cancer without a tensin switch 2015 [103] O K K M S E T K I H N A T Y G Y S I T Y Togashi Y, Shirakawa J -cell proliferation after a partial pancreatectomy is independent of irs-2 in mice 2014 [104] D D B M A G M F G T Wagner M, Kunsch S Transgenic overexpression of the oncofetal rna binding protein koc leads to remodeling of the exocrine pancreas Gastroenterology, 2003 [105] K S H M M P C M D P Walsh N, O’Donovan N Identification of pancreatic cancer invasion-related proteins by proteomic analysis 2009 58 [106] K S H M M P C M D P Walsh N, O’Donovan N Identification of pancreatic cancer invasion-related proteins by proteomic analysis 2009 [107] H Wang, P Lü, S Yue, L Chang, Y Li, H Zhao, W Li, and J Gao Combined use of wide-detector and adaptive statistical iterative reconstruction-v technique in abdominal ct with low radiation dose Zhonghua yi xue za zhi, 97(45) :3567–3572, 2017 [108] L Wang, Y Wang, and Q Chang Feature selection methods for big data bioinformatics : A survey from the search perspective Methods, 111 :21–31, 2016 [109] L Wang, Y Zhang, W Wang, Y Zhu, Y Chen, and B Tian Gemcitabine treatment induces endoplasmic reticular (er) stress and subsequently upregulates urokinase plasminogen activator (upa) to block mitochondrial-dependent apoptosis in panc-1 cancer stem-like cells (cscs) PloS one, 12(8) :e0184110, 2017 [110] T Y D C X Y T X H M Y Z J X Z B C S L Z Wang D, Zhu ZM Identfication of key mirnas in pancreatitis using bioinformatics analysis of microarray data 2016 [111] S S Wang F, Olson EM Role of derlin-1 protein in proteostasis regulation of atp-sensitive potassium channels 2012 [112] J Y Z Y H S L T J X X G Wang S, Zhu HY Adrenergic signaling mediates mechanical hyperalgesia through activation of p2x3 receptors in primary sensory neurons of rats with chronic pancreatitis 2015 [113] R S S B X Y H Y J M L J B M B S O D Wang Y, Kumar S Polyplex-mediated inhibition of chemokine receptor cxcr4 and chromatin-remodeling enzyme ncoa3 impedes pancreatic cancer progression and metastasis 2016 [114] B D G P H H V H D Warner SL, Muñoz RM Pdx-1-driven overexpression of aurora a kinase induces mild ductal dysplasia of pancreatic ducts near islets in transgenic mice 2014 [115] B F B S H U S J Welsch T, Keleg S Actinin-4 expression in primary and metastasized pancreatic ductal adenocarcinoma 2009 [116] H Xiang, Q Zhang, B Qi, X Tao, S Xia, H Song, J Qu, and D Shang Chinese herbal medicines attenuate acute pancreatitis : pharmacological activities and mechanisms Frontiers in pharmacology, :216, 2017 59 [117] S Xu, F Sun, L Ren, H Yang, N Tian, and S Peng Resveratrol controlled the fate of porcine pancreatic stem cells through the wnt/β-catenin signaling pathway mediated by sirt1 PloS one, 12(10) :e0187159, 2017 [118] B Xue, W Fu, and M Zhang Multi-objective Feature Selection in Classification : A Differential Evolution Approach In Proceedings of the 10th International Conference on Simulated Evolution and Learning - Volume 8886, pages 516–528, Dunedin, 2014 SpringerVerlag New York, Inc [119] B Xue, M Zhang, W N Browne, and X Yao A Survey on Evolutionary Computation Approaches to Feature Selection IEEE Transactions on Evolutionary Computation, 20(4) :606–626, 2016 [120] M M S K N Y M T O H F Y M T Yamagata T1, Tsuru T Genome organization of human 48-kda oligosaccharyltransferase (ddost) 1997 [121] Z Yan, J Gao, X Lv, W Yang, S Wen, H Tong, and C Tang Corrigendum to “quantitative evaluation and selection of reference genes for quantitative rt-pcr in mouse acute pancreatitis” BioMed research international, 2017, 2017 [122] N V R K L X T A P M Z J Yao G, Su X Polycystin-1 regulates actin cytoskeleton organization and directional cell migration through a novel pc1-pacsin 2-n-wasp complex 2014 [123] T H M T T T M H H T F M A T I M M F N T K Y E S U M Yoshimatsu G, Sakata N The co-transplantation of bone marrow derived mesenchymal stem cells reduced inflammation in intramuscular islet transplantation 2015 [124] M Yousef, S Jung, L C Showe, and M K Showe Recursive cluster elimination (rce) for classification and feature selection from gene expression data BMC bioinformatics, 8(1) :144, 2007 [125] M Zeng, S Chen, J Zhang, Z Wu, M Wang, R Jia, D Zhu, M Liu, K Sun, Q Yang, et al Molecular identification of goose (anser cygnoide) suppressor ubiquitin-specific protease 18 (usp18) and the effects of goose ifn and tmuv on its comparative transcripts Poultry science, 2017 [126] S X X L Y Y W L S A Zhao Y1, Xue Q1 First identification of the toxicity of microcystins on pancreatic islet function in humans and the involved potential biomarkers Environment science technology, 2016 60 [127] W C H X X C L K X J Zheng B, Yang L Curcumin analog l3 alleviates diabetic atherosclerosis by multiple effects European journal of pharmacology, 2016 [128] X Zhu, C Shi, Y Peng, L Yin, M Tu, Q Chen, C Hou, Q Li, and Y Miao Thymidine kinase silencing retards proliferative activity of pancreatic cancer cell via e2f1-tk1-p21 axis Cell proliferation, 2017 [129] Z Zhu, Y.-S Ong, and M Dash Markov blanket-embedded genetic algorithm for gene selection Pattern Recognition, 40(11) :3236–3248, 2007 [130] E Zitzler, K Deb, and L Thiele Comparison of multiobjective evolutionary algorithms : Empirical results Evolutionary computation, 8(2) :173–195, 2000 61 ... le cancer du pancréas en utilisant les algorithmes évolutifs Xác định chất điểu chỉnh ung thư tuyến tụy cách sử dụng thuật tốn tiến hóa Spécialité: Systèmes intelligents et multimédia Code: Programme... LỜI CAM ĐOAN Tơi cam đoan cơng trình nghiên cứu riêng Các số liệu, kết nêu Luận văn trung th ực ch ưa t ừng đ ược công bố cơng trình khác Các thơng tin trích d ẫn Lu ận văn rõ nguồn gốc Signature