colbyago1 commited on
Commit
aefec1b
·
verified ·
1 Parent(s): 86f0298

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -8
README.md CHANGED
@@ -8,7 +8,7 @@ tags:
8
  - chemistry
9
  dataset_summary: ML Application Curated Data from The Structural Antibody Database (SAbDab)
10
  pretty_name: The Curated SAbDab
11
- dataset_description: This dataset contains curated data from the SAbDab as of Mar 4, 2026. The SAbDab is a database of antibody structures including experimental details, antibody nomenclature, affinity data and sequence annotations. This data is filtered to include high quality, antigen-bound, and non-redundant antibody structures only. Non-redundant structures were obtained directly from the SAbDab database with query parameters Max. sequence identity 90 %, both paired VH/VL structures and unpaired, and both bound and unbound complexes
12
  acknowledgements: 'We kindly acknowledge the SAbDab team, RosettaCommons, and the following institutions: University of California, Los Angeles; University of Maryland; University of Oregon; University of Michigan; University of Pennsylvania; and the Wistar Institute'
13
  size_categories:
14
  - 10K<n<100K
@@ -129,13 +129,7 @@ which is a column oriented format that can be accessed directly, converted in to
129
  ## Dataset Details
130
 
131
  ### Dataset Description
132
- This dataset contains curated data from the SAbDab as of Mar 4, 2026. The
133
- SAbDab is a database of antibody structures including experimental details,
134
- antibody nomenclature, affinity data and sequence annotations. This data is
135
- filtered to include high quality, antigen-bound, and non-redundant antibody
136
- structures only. Non-redundant structures were obtained directly from the
137
- SAbDab database with query parameters Max. sequence identity 90 %, both paired
138
- VH/VL structures and unpaired, and both bound and unbound complexes
139
 
140
  - **Acknowledgements:**
141
  We kindly acknowledge the SAbDab team, RosettaCommons, and the following institutions: University of California, Los Angeles; University of Maryland; University of Oregon; University of Michigan; University of Pennsylvania; and the Wistar Institute.
 
8
  - chemistry
9
  dataset_summary: ML Application Curated Data from The Structural Antibody Database (SAbDab)
10
  pretty_name: The Curated SAbDab
11
+ dataset_description: This dataset contains curated data from the SAbDab as of Mar 4, 2026. The SAbDab is a database of antibody structures including experimental details, antibody nomenclature, affinity data and sequence annotations. Additional datasets include a ‘in_nr_set’ that was assembled by querying SAbDab with Max. sequence identity 90% and ‘curated_quality_dataset’ that was assembled by querying SAbDab with Max. sequence identity 90%, bound structures, resolution below 4.0 Å. We also performed multivariate stratification on "antigen_type", "heavy_species", "method", "scfv", "engineered", "light_ctype", "in_nr_set", "curated_quality_dataset" to achieve an 80/10/10 train/test/validation split
12
  acknowledgements: 'We kindly acknowledge the SAbDab team, RosettaCommons, and the following institutions: University of California, Los Angeles; University of Maryland; University of Oregon; University of Michigan; University of Pennsylvania; and the Wistar Institute'
13
  size_categories:
14
  - 10K<n<100K
 
129
  ## Dataset Details
130
 
131
  ### Dataset Description
132
+ This dataset contains curated data from the SAbDab as of Mar 4, 2026. The SAbDab is a database of antibody structures including experimental details, antibody nomenclature, affinity data and sequence annotations. Additional datasets include a ‘in_nr_set’ that was assembled by querying SAbDab with Max. sequence identity 90% and ‘curated_quality_dataset’ that was assembled by querying SAbDab with Max. sequence identity 90%, bound structures, resolution below 4.0 Å. We also performed multivariate stratification on "antigen_type", "heavy_species", "method", "scfv", "engineered", "light_ctype", "in_nr_set", "curated_quality_dataset" to achieve an 80/10/10 train/test/validation split
 
 
 
 
 
 
133
 
134
  - **Acknowledgements:**
135
  We kindly acknowledge the SAbDab team, RosettaCommons, and the following institutions: University of California, Los Angeles; University of Maryland; University of Oregon; University of Michigan; University of Pennsylvania; and the Wistar Institute.