Accessing Monoclonal Antibody Data

Workflow overview

Using the DataSpace app, the workflow of using the mAb grid is the following:

Navigate to the mAb Grid and browse the available mAb mixtures
Select the mAb mixtures that you’d like to investigate
Or filter rows by using columns:
- mAb/Mixture
- donor species
- isotype
- HXB2 location
- tiers
- clades
- viruses
Click “Neutralization Curves” or “IC50 Titer Heatmap” to visualize the mAb data
Click “Export CSV” or “Export Excel” to download the mAb data

DataSpaceR offers a similar interface:

Browse the mAb Grid by con$mabGridSummary
Select the mAb mixtures by filtering the mAb grid using any columns found in con$mabGrid using con$filterMabGrid()
Use con$getMab() to retrieve the mAb data

Browse the mAb Grid

You can browse the mAb Grid by calling the mabGridSummary field in the connection object:

library(DataSpaceR)
con <- connectDS()

knitr::kable(head(con$mabGridSummary))

mab_mixture	donor_species	isotype	hxb2_location	n_viruses	n_clades	n_tiers	geometric_mean_curve_ic50	n_studies
10-1074	human	IgG	Env	7	3	2	0.0213723	1
10E8	human	IgG3	gp160	227	11	7	0.4843333	2
10E8 V2.0	human			28	3	1	0.0031350	1
10E8 V2.0/iMab	human		gp160	13	8	2	0.0462897	1
10E8 V4.0	human			28	3	1	0.0024094	1
10E8 V4.0/iMab	human			119	12	6	0.0015396	1

This table is designed to mimic the mAb grid found in the app.

One can also access the unsummarized data from the mAb grid by calling con$mabGrid.

Filter the mAb grid

You can filter rows in the grid by specifying the values to keep in the columns found in the field con$mabGrid: mab_mixture, donor_species, isotype, hxb2_location, tiers, clades, viruses, and studies. filterMabGrid takes the column and the values and filters the underlying tables (private fields), and when you call the mabGridSummary or (which is actually an active binding), it returns the filtered grid with updated n_ columns and geometric_mean_curve_ic50.

# filter the grid by viruses
con$filterMabGrid(using = "virus", value = c("242-14", "Q23.17", "6535.3", "BaL.26", "DJ263.8"))

# filter the grid by donor species (llama)
con$filterMabGrid(using = "donor_species", value = "llama")

# check the updated grid
knitr::kable(con$mabGridSummary)

mab_mixture	donor_species	hxb2_location	n_viruses	n_clades	n_tiers	geometric_mean_curve_ic50	n_studies
11F1B	llama		4	2	1	NA	1
11F1F	llama		4	2	1	26.2961178	1
1H9	llama	Env	4	2	1	5.0898322	1
2B4F	llama		4	2	1	1.5242288	1
2H10	llama		2	2	1	NA	1
2H10/W100A	llama		2	2	1	NA	1
3E3	llama	gp160	3	3	1	0.9944945	1
4H73	llama		4	2	1	NA	1
5B10D	llama		4	2	1	NA	1
9B6B	llama		4	2	1	24.3643637	1
A14	llama	gp160	3	3	1	1.8444582	1
B21	llama	gp160	3	3	1	0.0936399	1
B9	llama	gp160	3	3	1	0.0386986	1
LAB5	llama		4	2	1	NA	1

Or we can use method chaining to call multiple filter methods and browse the grid. Method chaining is unique to R6 objects and related to the pipe. See Hadley Wickham’s Advanced R for more info.

con$resetMabGrid()
con$
  filterMabGrid(using = "virus", value = c("242-14", "Q23.17", "6535.3", "BaL.26", "DJ263.8"))$
  filterMabGrid(using = "donor_species", value = "llama")$
  mabGridSummary

Retrieve column values from the mAb grid

You can retrieve values from the grid by mab_mixture, donor_species, isotype, hxb2_location, tier, clade, virus, and studies, or any variables found in the mabGrid field in the connection object via data.table operations.

# retrieve available viruses in the filtered grid
con$mabGrid[, unique(virus)]
#> [1] "6535.3"  "Q23.17"  "DJ263.8" "BaL.26"  "242-14"

# retrieve available clades for 1H9 mAb mixture in the filtered grid
con$mabGrid[mab_mixture == "1H9", unique(clade)]
#> [1] "CRF02_AG" "B"

Create a DataSpaceMab object

After filtering the grid, you can create a DataSpaceMab object that contains the filtered mAb data.

mab <- con$getMab()
mab
#> <DataSpaceMab>
#>   URL: https://dataspace.cavd.org
#>   User: [email protected]
#>   Summary:
#>     - 3 studies
#>     - 14 mAb mixtures
#>     - 1 neutralization tiers
#>     - 3 clades
#>   Filters:
#>     - virus: 242-14, Q23.17, 6535.3, BaL.26, DJ263.8
#>     - mab_donor_species: llama

There are 6 public fields available in the DataSpaceMab object: studyAndMabs, mabs, nabMab, studies, assays, and variableDefinitions, and they are equivalent to the sheets in the excel file or the csv files you would download from the app via “Export Excel”/“Export CSV”.

knitr::kable(con$mabGridSummary)

mab_mixture	donor_species	hxb2_location	n_viruses	n_clades	n_tiers	geometric_mean_curve_ic50	n_studies
11F1B	llama		4	2	1	NA	1
11F1F	llama		4	2	1	26.2961178	1
1H9	llama	Env	4	2	1	5.0898322	1
2B4F	llama		4	2	1	1.5242288	1
2H10	llama		2	2	1	NA	1
2H10/W100A	llama		2	2	1	NA	1
3E3	llama	gp160	3	3	1	0.9944945	1
4H73	llama		4	2	1	NA	1
5B10D	llama		4	2	1	NA	1
9B6B	llama		4	2	1	24.3643637	1
A14	llama	gp160	3	3	1	1.8444582	1
B21	llama	gp160	3	3	1	0.0936399	1
B9	llama	gp160	3	3	1	0.0386986	1
LAB5	llama		4	2	1	NA	1

View metadata concerning the mAb object

There are several metadata fields that can be exported in the mAb object.

names(mab)
#>  [1] ".__enclos_env__"     "variableDefinitions" "assays"             
#>  [4] "studies"             "nabMab"              "mabs"               
#>  [7] "studyAndMabs"        "config"              "clone"              
#> [10] "getLanlMetadata"     "refresh"             "print"              
#> [13] "initialize"

DataSpaceR can also fetch and add metadata associated with downloaded mAbs via the getLanlMetadata method that is associated with the DataSpaceMab object.

mab$getLanlMetadata()

The LANL metadata can now be found at the mabs$lanl_metadata variable. This is a list column and its structure can very depending on what data LANL has collected.

mab$mabs[mab_name_std == "B9"]$lanl_metadata
#> [[1]]
#> [[1]]$epitopes
#>    accession alt_names      binding_type               cite   country dis
#> 1:        NA        NA <data.table[1x2]> <data.table[1x25]> <list[0]>  NA
#>    disrange donor epitope epitope_name hxb2_contig hxb2loc2end hxb2loc2start
#> 1:       NA    NA      NA           NA          NA          NA            NA
#>    hxb2locend hxb2locstart hxb2protein hxb2protein_id   id         immunogen
#> 1:         NA           NA       gp160             18 3219 <data.table[1x2]>
#>    in_catnap in_feature_db is_adcc   isotype           keyword mab_name
#> 1:      TRUE         FALSE      NA <list[0]> <data.table[4x2]>       B9
#>             modifydate neutralizing              note origlocend origlocstart
#> 1: 2018-06-01 13:59:55            L <data.table[2x4]>         NA           NA
#>    origprotein origprotein_id   patient           species strain subprotein
#> 1:          NA              1 <list[0]> <data.table[1x2]>     NA         NA
#>    subprotein_id           subtype table total_cite_count total_note_count
#> 1:             1 <data.table[2x2]>    ab                1                2
#>    vaccine_adjuvant vaccine_component    vaccine_strain      vaccine_type
#> 1:        <list[0]> <data.table[1x2]> <data.table[2x2]> <data.table[1x2]>
#> 
#> [[1]]$params
#> [[1]]$params$id
#> [1] "3219"
#> 
#> [[1]]$params$table
#> [1] "ab"
#> 
#> 
#> [[1]]$timestamp
#> [1] "2022-04-01 17:01:42z"
#> 
#> [[1]]$source
#> [1] "https://www.hiv.lanl.gov/mojo/immunology/api/v1/epitope/ab?id=3219"

Session information

sessionInfo()
#> R version 4.1.2 (2021-11-01)
#> Platform: x86_64-pc-linux-gnu (64-bit)
#> Running under: Ubuntu 18.04.5 LTS
#> 
#> Matrix products: default
#> BLAS:   /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.7.1
#> LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.7.1
#> 
#> locale:
#>  [1] LC_CTYPE=en_US.utf8       LC_NUMERIC=C             
#>  [3] LC_TIME=en_US.utf8        LC_COLLATE=en_US.utf8    
#>  [5] LC_MONETARY=en_US.utf8    LC_MESSAGES=en_US.utf8   
#>  [7] LC_PAPER=en_US.utf8       LC_NAME=C                
#>  [9] LC_ADDRESS=C              LC_TELEPHONE=C           
#> [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C      
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> other attached packages:
#> [1] data.table_1.14.2 DataSpaceR_0.7.5  knitr_1.37       
#> 
#> loaded via a namespace (and not attached):
#>  [1] Rcpp_1.0.8       digest_0.6.29    assertthat_0.2.1 R6_2.5.1        
#>  [5] jsonlite_1.8.0   magrittr_2.0.2   evaluate_0.15    highr_0.9       
#>  [9] httr_1.4.2       stringi_1.7.6    curl_4.3.2       tools_4.1.2     
#> [13] stringr_1.4.0    Rlabkey_2.8.3    xfun_0.29        compiler_4.1.2