Welcome to ctMethTracer
ctMethTracer (tissue-of-origin prediction using methylation profiles of cell-free DNA) is a web server that infers the proportions and the tissue-of-origin of tumor-derived cell-free DNA in a blood sample using DNA methylation data. To date, DNA methylation has been realized to be an ideal target for cancer diagnosis in clinical practice. The non-invasive hallmark of ctDNA methylation makes it a promising strategy for general cancer screen. Here, ctMethTracer attempts to provide a platform for the tissue-of-origin prediction by integrating published epigenomic data and algorithms. Currently, four probabilistic methods are provided in ctMethTracer (CancerLocator, random forest, logistic regression and support vector machine).
Step 1: select a file
Choose an example.
Download the example data you chose.
Upload your data
Upload your tumor-derived cell-free DNA methylation data (Hg19).
Input files can be created from the output of Bismark or BSMAP using the programs provided on the right.
Step 2: select model
CancerLocator uses a probabilistic method which simultaneously infers the proportions and the tissue-of-origin of tumor-derived cell-free DNA in a blood sample using genome-wide DNA methylation data.
Random forest is an ensemble learning method for classification that operate by constructing a multitude of decision trees at training time and outputting the class that is the mode of the classes of the individual trees.
Multinomial logistic regression is a model used to predict the probabilities of the different possible outcomes of a categorically distributed dependent variable, given a set of independent variables.
Support vector machines (SVMs) are supervised learning models with associated learning algorithms that analyze data used for classification.