0% Complete
صفحه اصلی
/
اولین کنگره بین المللی رویکردهای نوین سبک زندگی، پیشگیری و درمان سرطان
Prediction in protein-protein interaction networks in BRCA1, cancer cell using Ensemble method
نویسندگان :
Mahshid Sahraei (Iran University of Medical Sciences, Tehran, Iran)
کلمات کلیدی :
protein-protein° interaction،machine learning،canser،BRCA1،Bagging method
چکیده :
Protein-protein interaction (PPI) occurs as a result of high-specificity physical contacts between two or more protein molecules and causes biochemical events driven by interactions that include electrostatic forces, hydrogen bonding, and hydrophobic effects. On the other hand, protein interaction Protein plays a key role in predicting the target protein-protein function and the medicinal ability of molecules. The description and knowledge of protein-protein interaction networks helps a lot in the analysis of signal transduction pathways. all of which can lead to the construction of networks that predict all. Currently, knowing the function of the undiscovered protein-protein interaction (PPI) is one of the key issues for the development and progress of modern systems biology. With full knowledge and understanding of protein-protein interaction, it is possible to enter the bioinformatics science through which to predict possible links to find treatment methods, drug therapy, prognosis of chronic diseases, such as cancer, etc., both now and in the future Future tense earned. Considering that the main goal of this research is to extract the necessary knowledge from past experiences and/or find patterns in the data, therefore, we do this by machine learning, ML. so that, thereby, the algorithms perform the prediction process among the links. Since the structural knowledge of protein-protein interaction (PPI) relationships shows efficient information about kinetics, thermodynamics and molecular functions in the complex and defines its role in diseases. As far as it is possible to understand complex protein-protein°structures, °machine learning methods can be used. To discover these relationships, since the use of a basic method in machine learning can have a lower accuracy compared to our expectation level, therefore, the combined method of machine learning (Ensemble) with the Bagging method and the combination of two basic algorithms, Decision Tree and Random forest We used it by accident. The reason for using the Bagging algorithm is that the Bagging technique in machine learning, Bagging, provides information about the types of Bagging algorithms. Bagging is a powerful technique that helps reduce variance, which in turn prevents overfitting. Ensemble methods improve model accuracy by using a group of models. Which, when combined, work better than separate models when used separately. Taking the help of machine learning methods, instead of blindly examining all possible links, we can better find predictions based on the observed links and better focus on the links that are most likely to exist. The selected protein in this study is (BRCA1), which has eleven relationships with other proteins (TOPBP1-BABAM1-BABAM2-BARD1-BRCA1-MRE11-BRIP1-TP53-PALB2-FANCD2-ATM) and was selected from the STRING database. The purpose of this research is to increase the accuracy of the research to 100%, the implementation time (0 seconds), without any errors and the average error is 0.005, using the ensemble algorithm method. Also, the results of this study show that the use of Ensemble algorithm methods on methods of treating the disease. However, due to the experimental delay in solving abase data can be doctor's help in designing diagnostic and treatment process systems doctors help in designing diagnostic and treatment process systems.
لیست مقالات
ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 34.4.5
