Notes from the STEM Community Call on Thursday, April 7, 2016
On the call: Jamie, Kun, Matt, Nereyda, Chris, Werner, Judy
Next Community Call: Thursday, May 12th
1. STEM Milestone build 3.0.2, aka Java 8/Eclipse 4 build
*Up and tested
*Date for 3.0.3 to be scheduled for June (Stefan)
2. Feature requests
*Evolving disease model to address genome. Jamie is working w/ Matt and Kun on this.
*Extended shape file capabilities using a Pajek file. Chris has completed; is in and documented.
*Progress with travel model. Nereyda and Jamie to work together on April 8, will report in May.
*Zika. Kun is tracking; no decision yet as to how/whether to address.
*Followup on STEM workshop with Roslyn and Raymond. Completed.
*ISCB Fight against Ebola. Decided not to prepare a proposal.
*Bug 465981: Kamesh’s problem with dengue model. TO DO: Kun to verify.
*New bug: Running headless. Matt has done workaround. TO DO: Jamie will specify display in FHS window, will test; if it works, will document on wiki; if it doesn’t work, Philippine team will address.
*New bug [#465981?]: Batch experiments failing when transformation decorator is used. Influx of reports on news group from users, perhaps due to use w/ zika; bug real, but not a high priority.
5. Items from participants
*Jamie: Will be writing new papers on evolving disease model this summer w/ UC Davis microbiologists. USDA has genome project identifying 5-6 SNFs in salmonella to use in tracking; SNFs in bacteria aren’t stable; when a mutation occurs, the genome changes and there is a quasi species. Goal is to combine epidemiology and genomics. Builds on earlier work Chris did w/ transformation decorators for slaughter houses; also Matt’s model generator for discontinuous transformation. Will model cross strain immunity, change in genetic distance. Given food safety focus, is this of interest to BfR? TO DO: Chris will speak with Matthias.
*Jamie: Located multiple papers using STEM on Google. TO DO: Judy will go through, find any not cited on STEM wiki and add to listing there.
*Chris: Institute in Berlin responsible for infectious diseases in humans has hired a new mathematical modeler, Dirk Baughman, previously at Northeastern U in Boston, where they used another modeler. Baughman finds too many models that too complicated and make it hard to draw conclusions. Jamie: Would like introduction; Kun’s paper on three dengue models deemed one to be over determined; multiple approaches to modeling exist.
*Werner: To attend Eclipse Science Working Group June 10. Jamie: If any Eclipse software engineers interested, could use their help on fixing things that broke when STEM moved from Eclipse 3 to Eclipse 4; interested in WG if we can retain our URL, someone from IBM already involved. Jamie: Reached out to steering committee’s Matthew Goering on Linked In. TO DO: Chris will check with Matthias as to BfR’s interest in a meeting in Berlin, will let Werner know.
*Jamie and Kun: Want to talk with Matthias about projects, US funding opportunity on food safety (paper IBM and BfR co-authored). TO DO: Chris to let Matthias know.