We get a range of zoonotic disease modelers in the course every year. Classic example types would be:
Rabies, with a spatial migration network
Not sure how much data exists in this field. Original modeling studies back in the 1990s worked with a single animal type (typically foxes) and placed a hypothetical extreme value distribution on the spatial "jump" process distribution, with seasonal forcing.
Things to represent: multi-animal transmission process (dogs, bats, humans?)
Foot & Mouth with a transport network
Interesting qs here include:
- How to define the nodes, links and layers --
- How to define the environmental force of infection
- How to treat the transport network -- is it totally fixed or is there some random process that changes the network?
Are the links fixed?
Are the acts fixed in time? (e.g., every Friday animals are sent from farm A to farm B)
Are the same number of animals shipped each time?
Roosting colonies
This is an interesting example, b/c this paper uses EpiModel to study spread of infection among roosting bats did not use ERGM to model/test differences in the bat contact networks, but they could have, and the fact that they plug the estimates into EpiModel makes that clear. So this would make for a nice gallery example where we fit the bat data instead with ERGM and compare the results.
The main reference it relies on for these non-ERGM methods is this:
Constructing, conducting and interpreting animal social network analysis
Farine, DR and Whitehead, H
Sep 2015
JOURNAL OF ANIMAL ECOLOGY 84(5), pp.1144-1163
Which doesn't reference any of the ERGM papers and appears to be unfamiliar with their development.
Bottom line: there is a huge literature on animal networks in EEB that is unaware of the ERGM/TERGM framework. We're getting some of those students every year. Would be great to either work with them or give them the tools to bring back to their communities.
We get a range of zoonotic disease modelers in the course every year. Classic example types would be:
Rabies, with a spatial migration network
Not sure how much data exists in this field. Original modeling studies back in the 1990s worked with a single animal type (typically foxes) and placed a hypothetical extreme value distribution on the spatial "jump" process distribution, with seasonal forcing.
Things to represent: multi-animal transmission process (dogs, bats, humans?)
Foot & Mouth with a transport network
Interesting qs here include:
Are the links fixed?
Are the acts fixed in time? (e.g., every Friday animals are sent from farm A to farm B)
Are the same number of animals shipped each time?
Roosting colonies
This is an interesting example, b/c this paper uses EpiModel to study spread of infection among roosting bats did not use ERGM to model/test differences in the bat contact networks, but they could have, and the fact that they plug the estimates into EpiModel makes that clear. So this would make for a nice gallery example where we fit the bat data instead with ERGM and compare the results.
The main reference it relies on for these non-ERGM methods is this:
Constructing, conducting and interpreting animal social network analysis
Farine, DR and Whitehead, H
Sep 2015
JOURNAL OF ANIMAL ECOLOGY 84(5), pp.1144-1163
Which doesn't reference any of the ERGM papers and appears to be unfamiliar with their development.
Bottom line: there is a huge literature on animal networks in EEB that is unaware of the ERGM/TERGM framework. We're getting some of those students every year. Would be great to either work with them or give them the tools to bring back to their communities.