Community
Participate
Working Groups
Using the Trigger, Modifier and Predicate files, I can create changes in certain disease parameters GLOBALLY based on time. It would be nice to be able to tie the modifications in disease parameters to location as well as time. For example, if we are using a reduced transmission rate to simulate school closures at day 10, how can we specify that only certain schools are closed? We would like to be able to apply the reduced transmission rate only to specific locations where the schools are closed and then let effects trickle out from there.
Stefan, Can you take this one?
I'm proposing adding a location attribute to all disease models to be consistent with how population models work. Changing target milestone 1.3.1.
This is an old feature request I propose we can discuss on this weeks call. My suggestion is to use a control graph and the control graph generator to attach labels to regions where we want to reduce the transmission rate. Adding a location to a disease model quickly gets tricky implications for logging, map view, etc.
Hi, I need to work on a research by using STEM with sublocations. This seems something really helpful for us. I'd like to know is there any progress on this request?
Hi, thanks, can you tell me a little more about what you are trying to do? We have models in STEM where the disease parameters can vary from region to region, but typically it's due to external factors like temperature, rainfall, latitude etc.
I need a model that includes the interaction between individuals who are in the same location to calculate some location based transmission rate. As far as I understood STEM provides models only based on regions of any country. I can go to the depth level 0,1,2 ( i.e. US-> New York-> NYC). But I need to work on the buildings, because we're trying to generate a model where each location produces a different transmission rate and individuals move a lot between those locations.
Okay. Also please see my response to your newsgroup post. It sounds like your disease model is different from any other we provide in STEM. We'd be happy to help you create a new disease model if you like. Here's some wiki links to the new model generator function in STEM that should help you get started: http://wiki.eclipse.org/STEM_Model_Generator including installation instructions: http://wiki.eclipse.org/STEM_Model_Generator/Install
Thank you Stefan. I really want to use STEM if my model is doable by using Model Generator. Actually if it is doable by STEM! Because it seems I need to use the nodes to represent each individual and the edges for their communications, co-existing in the same location for a while. Each individual will have different infectious rate in different locations and will interact each other by traveling through the locations. Even if I succeed to generate my own network by using pajek, does this structure seem to be able to be generated by STEM model generator to you? Sorry for complicated description of my thoughts but what I need is exactly using the individuals as nodes and I'm not sure STEM model generator would allow me to do that. In STEM, it looks like the graph is the geographical graph that represents the connection of different location.Will I be able to generate a human graph via pajek/stem interaction? I also wonder that how the STEM simulate the disease transmission between humans? How does STEM simulation know the number of population in each nodes (locations)? I know I have lots of questions but your answers will be really appreciated. Thanks a lot. Sanem
