Xu GJ, Kula T, Qikai Xu Q, et al. Comprehensive serological profiling of human populations using a synthetic human virome. Science. 2015:348(6239):1105 DOI: 10.1126/science.aaa0698
The VirScan system can measure a subject's exposure to all known human viruses.
For a patient with autoimmune disease, the system will be able to reveal the entire history of viral exposure. This will enable population studies that find any associations with a particular viral infection for every rheumatological disease. And it only costs about $25 per blood sample.
So this can be used to study viral interactions in diseases like type 1 diabetes, inflammatory bowel disease, and asthma, the authors say.
The system identifies an antibody to a virus fragment in the patient’s blood, and synthesizes the DNA to that fragment. A DNA microarray can then identify the virus fragment.
The group, led by Stephen Elledge at Harvard University, started with the reference protein sequences of all viruses reported to infect humans in the UniProt database. That was 206 viruses and over 1,000 strains. They then synthesized the corresponding DNA sequences. That came to 93,904 200-mer oligonucleotides, encoding 56-residue peptide tiles, with 28-residue overlaps. They inserted those DNA sequences into T7 phage vectors, which synthesized the viral protein back again and displayed it on their surface. The blood sample is mixed with those T7 phage vectors, and each antibody in the sample binds to the phage that is expressing its protein. The antibodies are immobilized, and the unbound phages are washed away. That leaves phages bound to an antibody, containing DNA for the antigen to that antibody.
After screening 569 donors from four continents, they detected antibodies to an average of 10 viral species per person, and 84 species in two individuals.
The limitations to the technology are that they only use published viruses, they use statistical methods of detection, and some immune reactions decline over time, below the threshold of the test. This may explain why they don’t report the expected prevalence of common viruses, nor the expected vaccine responses.