Add example BAMA dataset from DataSpaceR

[lemireg]

Chose an example dataset from DataSpaceR with a placebo group for use in example BAMA vignettes.

Description of data located here:
https://dataspace.cavd.org/cds/CAVD/app.view#learn/learn/Study/label=CAVD%20579

[asatofh]

@lemireg this is an incredible start, thanks!! A few things I like about this dataset - multiple groups / timepoints / antigens, and some decent responses. this would be enough to generate some response rate / magnitude and maybe mag-breadth (especially a panel, though I'm not sure they did).

The main issue is that it's missing differential binding data, which I think we need in a template dataset. The one publicly available dataset I know of with diff bind data is from IAVI C101. https://iavi.figshare.com/articles/dataset/Data_set_for_Precise_targeting_of_HIV_broadly_neutralizing_antibody_precursors_in_humans_/28873295?file=54542075
This repository contains BAMA and NAB data, both with differential KO data, which could be really useful to have in VISCtemplates. The data is wide, so not our current pdata format (though maybe our future format?). Take a look at this and let me know what you think. I want to leverage the work you've already done, but I also want to make sure whatever you write can be used long term!

[lemireg]

@bryanmayer :

• Add datahash to tests
• Add a sentence to Rd describing study
• Move to new R script: cds_dataset579

[lemireg]

@lemireg this is an incredible start, thanks!! A few things I like about this dataset - multiple groups / timepoints / antigens, and some decent responses. this would be enough to generate some response rate / magnitude and maybe mag-breadth (especially a panel, though I'm not sure they did).

The main issue is that it's missing differential binding data, which I think we need in a template dataset. The one publicly available dataset I know of with diff bind data is from IAVI C101. https://iavi.figshare.com/articles/dataset/Data_set_for_Precise_targeting_of_HIV_broadly_neutralizing_antibody_precursors_in_humans_/28873295?file=54542075 This repository contains BAMA and NAB data, both with differential KO data, which could be really useful to have in VISCtemplates. The data is wide, so not our current pdata format (though maybe our future format?). Take a look at this and let me know what you think. I want to leverage the work you've already done, but I also want to make sure whatever you write can be used long term!

@asatofh Looking at the BAMA.csv here, the diffBind column is empty. @drienna do you know who made the publication dataset for C101?

[lemireg]

@lemireg this is an incredible start, thanks!! A few things I like about this dataset - multiple groups / timepoints / antigens, and some decent responses. this would be enough to generate some response rate / magnitude and maybe mag-breadth (especially a panel, though I'm not sure they did).
The main issue is that it's missing differential binding data, which I think we need in a template dataset. The one publicly available dataset I know of with diff bind data is from IAVI C101. https://iavi.figshare.com/articles/dataset/Data_set_for_Precise_targeting_of_HIV_broadly_neutralizing_antibody_precursors_in_humans_/28873295?file=54542075 This repository contains BAMA and NAB data, both with differential KO data, which could be really useful to have in VISCtemplates. The data is wide, so not our current pdata format (though maybe our future format?). Take a look at this and let me know what you think. I want to leverage the work you've already done, but I also want to make sure whatever you write can be used long term!

@asatofh Looking at the BAMA.csv here, the diffBind column is empty. @drienna do you know who made the publication dataset for C101?

Per the conversation on 3/16/2026 with Alicia, Nazmus and Drienna, these data do not look right (antigen_ref_diffBind are BG, ratio_delta and ratio_fi_bkgd are always integers and equivalent, etc.). Drienna will follow up with IAVI.

[lemireg]

Per VISCtemplates group conversation on 3/19/2026, we will split out the example data in VISCfunctions which is currently used for testing, not to be also used as example data in vignettes. To generate example BAMA for use in vignettes we'll need:

• specifications for an example BAMA dataset. Questions include: which columns need to be included, which readouts (netMFI? responses? differential binding? if so, using what method(s)), how many antigens (if we're including magnitude breadth), which/how many groups (placebo? more than one treatment group?), and the format of the dataset (long? wide?)
• is the target to keep example datasets up to date with current VISC pdata standards? decide on goal and whether VISC has resources to upkeep based on the goal
• to make decisions about where these data will live (if anywhere): in it's own package (ex: VISCdata), in VISCfunctions? in VISCtemplates, or simply read-in/simulation as a part of the vignette itself