A clear prediction of lymph-node metastasis in certain cancers is important for targeted clinical interventions that allow favorable prognosis for patients. The webserver LNMpredictor was designed for forecasting lymph-node metastasis in cancers. The support vector machine (SVM) algorithm was implemented as individual classifying models based on different types of molecular expression profiles, including mRNA, microRNA (miRNA), and long non-coding RNA (lncRNA). The current version of LNMpredictor supports a spectrum of cancers, including Bladder urothelial carcinoma (BLCA), Breast invasive carcinoma (BRCA), Cervical and endocervical cancers (CESC), Colon adenocarcinoma (COAD), Kidney renal clear cell Carcinoma (KIRC), Lung adenocarcinoma (LUAD), Lung squamous cell carcinoma (LUSC), Pancreatic adenocarcinoma (PAAD), and Rectum adenocarcinoma (READ). With the submitted expression profile data of ~100 samples, LMNpredictor will take several seconds to finish the prediction and present the results in a webpage for researchers.
A text file documenting a type of expression profile (mRNA, miRNA, or lncRNA) for a number of samples should be firstly prepared, where gene expression quantification of them need to be used as data preprocessing by users. With the data, molecular profile types (mRNA, miRNA, and lncRNA), cancer types (9 types described above) should be sequentially selected prior to uploading the data (Browse -> Local disk -> Select target file -> Open). When clicking the "Predict" button, the results will be returned for users in several seconds.