How to handle pooled samples
Although PHAROS is designed for granular (animal-level) data, pooled testing is still a welcome data source.
Researchers who are pooling samples from multiple animals can present that information one of two ways:
Option 1: One-to-many sample-animal relationship
Suppose a researcher swabs three marmosets, and pools the three swabs into a single test for herpesviruses---but still wants to record metadata on individual animals. They might represent the single positive test result as:
Sample ID | Animal ID | Host species | Organism sex | Detection outcome | Pathogen | GenBank accession | |
---|---|---|---|---|---|---|---|
1 | CApool1 | Calla01 | Callithrix aurita | Female | Positive | Macacine alphaherpesvirus 1 | GB8675309 |
2 | CApool1 | Calla002 | Callithrix aurita | Male | Positive | Macacine alphaherpesvirus 1 | GB8675309 |
3 | CApool1 | Calla003 | Callithrix aurita | Female | Positive | Macacine alphaherpesvirus 1 | GB8675309 |
In this format, the metadata for each animal is given on a separate row, but attached to the "Sample ID" and "GenBank accession".
Option 2: One-to-none sample-animal relationship
Suppose a researcher pools several dozen mosquitoes from a trap and tests that sample for West Nile virus.
Sample ID | Animal ID | Host species | Detection outcome | Pathogen | GenBank accession | Detection comments |
---|---|---|---|---|---|---|
Tuscon1 | Aedes aegypti | Positive | West Nile virus | GB1738 | Pooled mosquito traps |
In this format, individual animals are not identified, and the finest resolution of the data is at the trap (= sample) level.