We argue for a more sweeping reappraisal of the term pathogen than Arturo Casadevall and Liise-anne Pirofski propose (Nature 516, 165–166; 2014). This should take in not just microbes, but the wider 'exposome' and recent discoveries in infection and immunity research.

A term is needed that encompasses sequences from the environment — intrinsic or extrinsic — that impart pathogenic or benign information to eukaryotic immune receptors. For example, T-cell receptors that recognize autoantigen and microbial sequences can be triggered by related peptide sequences from diverse sources in the environment (M. E. Birnbaum et al. Cell 157, 1073–1087; 2014).

Substituting 'microbial immunogen' for 'pathogen' would not account for microbiota sequences that instruct immune development rather than elicit protective immunity. We suggest instead the term perceptogen (microbial or environmental) to cover protein sequences that affect the body's range of reactions after perception by its immune receptors.

As the writer Aldous Huxley remarked: “There are things known and there are things unknown and in between are the doors of perception.”