Название: Machine Learning Analysis of QPCR Data Using R
Автор: Luigi Marongiu
Издательство: Nova Science Publishers
Год: 2022
Страниц: 150
Язык: английский
Формат: pdf (true)
Размер: 18.4 MB

The quantitative polymerase chain reaction (qPCR) is a versatile and popular assay for quantifying nucleic acids. With the recent expansion of the number of reactions per assay, there is a need for an accurate method to report the data suitable for automation. This book will describe such a method, based on Machine Learning (ML) analysis, and implement it with publicly available tools. This book is intended for researchers and will provide a detailed introduction to the programming language R, including references for the most common functions. This book will provide an advanced strategy for the objective analysis of qPCR data suitable for experts in the field and an introduction to qPCR and computational analysis for students.

Support-vector machines (SVM) are Supervised Learning models used for classification and regression analysis as part of Machine Learning algorithms. SVM was developed at AT&T Bell Laboratories by Vladimir Vapnik and colleagues between 1960s and the 1990s. SVM, in essence, finds a separation line (‘maximum margin hyperplane’ or simply hyperplane) between groups of samples known as items or elements. Linearly separable groups are non-overlapping groups. The SVM algorithm finds the hyperplane with the greatest degree of separation between the groups, meaning that the least number of elements from one group will be re-classified in the other group. Instead of using all elements from each group, SVM uses a subset of items closest to the hyperplane, which reduces the procedure’s complexity and computational requirements. When the groups are not linearly separable, SVM applies ‘margins’: some of the elements in proximity of the hyperplane are allowed to be clustered in the opposite group.