Metabolomics, proteomics, and genomics (in general omics) hold the promise as a new technology to diagnose highly heterogeneous diseases. Conventionally, omics data analysis for diagnosis is done using various statistical and machine learning based classification methods. However, it remains unknown if deep neural network, a class of increasingly popular machine learning methods, is suitable to classify omics data. Here we use omics data to test the accuracies of feedforward networks, a deep learning (DL) framework, as well as five widely used machine learning models, namely random forest (RF), support vector machines (SVM), linear discriminant analysis (LDA), prediction analysis for microarrays (PAM), and generalized linear models (GLM) to predict some very important diseases. DL framework resulted in higher predictive power in classifying cases/controls, compared to the other five machine learning algorithms. Some of the diseases that we work to predict are pancreatic cancer, cervical cancer, autism, down syndrome, cerebral palsy, pediatric concussion, miscarriage and Alzheimer disease.