Description: Single cell RNA-sequencing (scRNA-seq) is a transformative technology that is rapidly advancing our understanding of biology and the way we do biomedical research. However, the technical progress brings along cogent challenges. In most scRNA-seq studies, the data is extremely sparse and noisy, hindering downstream analyses. To address this problem, we had previously developed a Single-cell Analysis Via Expression Recovery (SAVER) method for gene expression denoising and imputation. We further incorporated a deep count autoencoder network into our algorithm, improving accuracy. Our algorithm, SAVERX, allows for high quality scRNA-seq data imputation across a variety of settings, and can greatly enhance biological discovery. To make it widely available to the scientific research community, we propose developing a user-friendly online platform where any researcher can upload their data and retrieve SAVERX-denoised data. For optimum performance, implementation and maintenance of such a web-tool would require high-performance clusters, and cloud data processing software and resources. Our algorithm utilizes parts of existing libraries developed in Python, Keras and R, all software programs that are available through XSEDE. Thus, an XSEDE-based platform would not only substantially accelerate the dissemination of our algorithm, but also be an ideal scaffold to support this large-scale scientific endeavor