Cytokines are small proteins that mediate signalling among immune and non-immune cells, and they trigger a range of cellular activity, such as proliferation, activation and killing. Over many decades, immunologists have described countless associations between cell types and the cytokines they produce or sense, but many of these findings, although published, are difficult to access. Associations may have been discovered in a particular disease context or cell type, or uncovered as part of a larger study and thus not corroborated or expanded. Work from Shai Shen-Orr and colleagues, published in Nature Biotechnology, aims to unearth these connections and provide a useful resource for enabling new discoveries. The researchers developed a computational tool that mines PubMed data and connects cell types to cytokines and diseases. The text-mining tool, called immuneXpresso, was used to identify connections between 340 cell types and 140 cytokines across thousands of diseases. Shen-Orr and colleagues showed they could corroborate known interactions and discover previously unappreciated connections worthy of further investigation. The resource is openly available and can be accessed here.