Changes between Version 8 and Version 9 of NIAID2022


Ignore:
Timestamp:
Aug 4, 2022, 2:02:16 PM (3 years ago)
Author:
Elaine Meng
Comment:

--

Legend:

Unmodified
Added
Removed
Modified

  • NIAID2022

    v8 v9  
    5959= Meeting Minutes =
    6060
     61== 8/18/2022 ==
     62
    6163== 8/4/2022 ==
    6264
    63 Phil, Meghan, Michal, Darrell, Philip; Zach, Eric, Elaine, TomG, Scooter
     65Phil, Meghan, Michal, Darrell, Philip; Zach, Eric, Elaine, TomG, Greg, Scooter
    6466
    6567Phil: what's going on with ChimeraX scripts for NIH3D?  Michal:
    … …  
    6870TomG: saw check-ins related to DICOM in pipeline code, are there issues we can help with? Michal: not that I know of, got pixel spacing, and will let you know if other issues arise. Scooter and Zach are setting up meetings with some UCSF radiology folks to get feedback on visualizing medical images. 
    6971
    70 Meghan, Phil: would be interested if they have any feedback on NIH3D as well, and if they're using AR/VR.  Meghan: we had visitors using AR, Hololens 2. TomG: we don't have a Hololens, would be interested in trying molecular data. Phil: we were working with a company Enklu to put a molecule in a scene for the Hololens 2. Static model was fine but trying to move it was very stuttery. Could have been due to the overhead of the Enklu environment. Meghan: it was "no code" but had a rather high learning curve.
    71 
    72 potential topics: 
    73 - Change Chain ID tool
    74 - how to get their pipeline to install the presets bundle in a reliable fashion
    75 - progress on Dock Prep
     72Meghan, Phil: would be interested if they have any feedback on NIH3D as well, and if they're using AR/VR.  Meghan: we had visitors using AR, Hololens 2. TomG: we don't have a Hololens, would be interested in trying molecular data. Phil: we were working with a company Enklu to put a molecule in a scene for the Hololens 2. Static model was fine but trying to move it was very stuttery. Could have been due to the overhead of the Enklu environment. Meghan: it was "no code" but had a rather high learning curve.  One limitation of Hololens is restricted field of view. Phil: had to play around with room lighting to see model clearly, rather than washed out.  Scooter: they are relatively expensive.  TomG: AR seems good for Jacobson group use case, where people discuss a few different structures without a lot of VR manipulation.
     73
     74From Zoom chat:
     75- Darrell: AR company I recently found out about: https://varjo.com/products/xr-3/
     76- Meghan: see https://varjo.com/products/xr-3/
     77
     78Meghan: we will be setting up a lot of hardware in Uganda, need to get more base stations. They keep failing, maybe were kept turned on continuously for too long. Even a used one is $270 on Amazon. Darrell: maybe should get on/off switch. TomG: maybe is better to leave them on rather than a lot of power cycling. But dust, power surges, temp fluctuations could all help to do them in. There are other ways to turn off, say via SteamVR.
     79
     80ChimeraX stuff.
     81Eric: there is a Change Chain ID tool. Have made progress on
     82Dock Prep, but it's very complicated and not there yet. Could be a couple more months.  Scooter: we are thinking about what features should be in version 1.5.
     83
     84TomG: Latest !AlphaFold DB release has >200 million structures. Can fetch a specific ID quickly but searching all the sequences takes a while.  Looking at RNA model database, and another method, Esmfold https://chowdera.com/2022/207/202207262300487520.html\
     85
     86Elaine: this paper presents AlphaFold2 performance on cryoEM challenge datasets https://www.sciencedirect.com/science/article/pii/S2589004222007672
     87
     88Meghan: we're also thinking about network visualization tools for our Uganda collaborations
    7689
    7790== 7/7/2022 ==