Cancer type; could it be identified correctly to an acceptable degree
- سال انتشار: 1391
- محل انتشار: هشتمین کنگره بین المللی سرطان پستان
- کد COI اختصاصی: ICBCMED08_336
- زبان مقاله: انگلیسی
- تعداد مشاهده: 400
نویسندگان
Isfahan UMS
چکیده
Introduction: The burden of cancer, especially breast cancer, urgesimprovements in its early detection to prevent incidence and costly treatments.DNA microarray is a new technology developed for nearly two decades and hasbecome a prominent area for researchers to focus on. It studies arrayed series ofthousands of microscopic spots of DNA oligonucleotides, called features,spotted on a chip by a robot. This technology is used to investigate expressionlevels of thousands of genes simultaneously and detecting the patterns of highlyexpressed genes in a disease. In this context, selecting fewer but the mostinformative genes to diagnose and predict the outcome of a disease is achallenging argument for researchers. Since its development in early 1990s,various methods have been used in microarray literature to approach this goal.Two of the most widely used statistical methods are clustering andclassification. In this study we combined different kinds of these methods toclassify and allocate a new breast cancer patient to one of apocrine, luminal, andbasal.Material And Methods:We used Hierarchical clustering technique to clustergenes and Support Vector Machine (SVM), Naive Bayes (NB), and decisiontree (C4.5) out of available classification methods predict to the class of apatient on the basis of their genome-wide expression levels. Before introducingto the algorithm to learn the patterns, the data was ranked by RF, Chi2, and IGmethods. The data on 49 patients with different types of breast cancer was usedto assess the performance of each classification method.Results: For IG, SVM and NB were the best classification methods with 100percent precision based on only13 and 3 genes, respectively. C4.5 was the bestone for RF and Chi2 ranking methods with 3 predicting genes.Conclusion: The performance of the machine learning methods may bedifferent on different data. But undoubtedly, using novel technologies ispreferable to traditional screening methods. On the other hand, microarrayscould be useful in detecting new patterns of expression in a certain type of adisease. It could be a useful monitoring tool for impacts of pharmaceuticaltherapies on expression levels of genes under studyکلیدواژه ها
مقالات مرتبط جدید
اطلاعات بیشتر در مورد COI
COI مخفف عبارت CIVILICA Object Identifier به معنی شناسه سیویلیکا برای اسناد است. COI کدی است که مطابق محل انتشار، به مقالات کنفرانسها و ژورنالهای داخل کشور به هنگام نمایه سازی بر روی پایگاه استنادی سیویلیکا اختصاص می یابد.
کد COI به مفهوم کد ملی اسناد نمایه شده در سیویلیکا است و کدی یکتا و ثابت است و به همین دلیل همواره قابلیت استناد و پیگیری دارد.