MALER: a web server to build, evaluate, and apply machine learning models for numerical biomedical data analysis

The explosive growth of numerical biomedical data poses a challenge in uncovering meaningful insights within from vast omics and clinical data. In recent years, machine learning has emerged as a powerful tool for processing and dissecting numerical biomedical data, making it a popular choice for addressing analytical challenges in biomedical research. Nevertheless, the intricate algorithms and complex optimization processes of machine learning model frameworks have deterred numerous non-machine learning experts. To address the need for comprehensive signatures exploration in biomedical field, we have developed Maler (http://www.inbirg.com/maler/home), a platform designed to facilitate the effortless and efficient application of machine learning models to high-dimensional biomedical data for bioscientists. Maler primarily offers four machine learning modes and a variety of machine learning algorithms to accommodate diverse analysis requirements. Simultaneously, it employs multiple feature selection methods to acquire relevant features and selects the optimal combinations of features based on machine learning models to achieve the best predictive outcomes. Maler is expected to provide a user-friendly and efficient platform for marker discovery, catering to the needs of biologists and clinical experts.