Updated with the benchmark details

README.md

@@ -17,47 +17,6 @@ dataset_info:
   splits:
   - name: train
     num_examples: 24700
-- config_name: CATS
-  features:
-  - name: Index 
-    dtype: string
-  - name: Drug
-    dtype: string
-  - name: Target
-    dtype: string
-  - name: Y
-    dtype: float32
-  splits:
-  - name: train
-    num_examples: 393
-- config_name: HIF2A
-  features:
-  - name: Index 
-    dtype: string
-  - name: Y
-    dtype: float32
-  - name: Drug
-    dtype: string
-  - name: Target
-    dtype: string
-  splits:
-  - name: train
-    num_examples: 37
-- config_name: HSP90
-  features:
-  - name: Index
-    dtype: string
-  - name: Drug
-    dtype: string
-  - name: IC50 (nM)
-    dtype: float32
-  - name: Target
-    dtype: string
-  - name: Y
-    dtype: float32
-  splits:
-  - name: train
-    num_examples: 147
 - config_name: LeakyPDB
   features:
   - name: Index 
@@ -95,33 +54,46 @@ dataset_info:
   splits:
   - name: train
     num_examples: 19443
-- config_name: MCL1
+- config_name: Mpro
   features:
   - name: Index 
     dtype: string
-  - name: Y
-    dtype: float32
   - name: Drug
     dtype: string
+  - name: Y
+    dtype: float32
   - name: Target
     dtype: string
   splits:
   - name: train
-    num_examples: 25
-- config_name: Mpro
+    num_examples: 2062
+- config_name: USP7
   features:
   - name: Index 
     dtype: string
+  - name: Y
+    dtype: float32
   - name: Drug
     dtype: string
+  - name: Target
+    dtype: string
+  splits:
+  - name: train
+    num_examples: 1799
+- config_name: MCL1
+  features:
+  - name: Index 
+    dtype: string
   - name: Y
     dtype: float32
+  - name: Drug
+    dtype: string
   - name: Target
     dtype: string
   splits:
   - name: train
-    num_examples: 2062
-- config_name: SYK
+    num_examples: 25
+- config_name: HIF2A
   features:
   - name: Index 
     dtype: string
@@ -133,8 +105,8 @@ dataset_info:
     dtype: string
   splits:
   - name: train
-    num_examples: 44
-- config_name: USP7
+    num_examples: 37
+- config_name: SYK
   features:
   - name: Index 
     dtype: string
@@ -146,48 +118,44 @@ dataset_info:
     dtype: string
   splits:
   - name: train
-    num_examples: 1799
+    num_examples: 44
+
 configs:
 - config_name: BindingDB_filtered
   data_files:
   - split: train
     path: BindingDB_filtered/train/data-*
-- config_name: CATS
-  data_files:
-  - split: train
-    path: CATS/train/data-*
-- config_name: HIF2A
+- config_name: LeakyPDB
   data_files:
   - split: train
-    path: HIF2A/train/data-*
-- config_name: HSP90
+    path: LeakyPDB/train/data-*
+- config_name: Mpro
   data_files:
   - split: train
-    path: HSP90/train/data-*
-- config_name: LeakyPDB
+    path: Mpro/train/data-*
+- config_name: USP7
   data_files:
   - split: train
-    path: LeakyPDB/train/data-*
+    path: USP7/train/data-*
 - config_name: MCL1
   data_files:
   - split: train
     path: MCL1/train/data-*
-- config_name: Mpro
+- config_name: HIF2A
   data_files:
   - split: train
-    path: Mpro/train/data-*
+    path: HIF2A/train/data-*
 - config_name: SYK
   data_files:
   - split: train
     path: SYK/train/data-*
-- config_name: USP7
-  data_files:
-  - split: train
-    path: USP7/train/data-*
+
 license: cc-by-4.0
 pretty_name: BALM-Benchmark
 tags:
   - chemistry
+  - deep learning
+  - protein-ligand binding affinity
   - biology
 size_categories:
   - 10K<n<100K
@@ -197,14 +165,31 @@ size_categories:
 
 <!-- Provide a quick summary of the dataset. -->
 
-BALM-Benchmark is a comprehensive benchmark suite that combines multiple seminal binding affinity prediction datasets in one place.
-...
+**BALM-Benchmark** is a curated collection of datasets designed to advance machine learning and deep learning model research for protein-ligand binding affinity prediction. This benchmark consolidates several key datasets including BindingDB, LP-PDBBind, and specific protein-ligand systems like USP7, MPro, SYK, HIF2A, and MCL1, each chosen for its distinct data characteristics and evaluation. 
+
+This dataset collection has been refined and standardized, making it readily accessible for deep learning model training and testing on [Hugging Face](https://huggingface.co/datasets/BALM/BALM-benchmark), providing a structured foundation for advancements in target-based drug discovery.
+
 
 ## Dataset Details
 
-### Dataset Description
 
-<!-- Provide a longer summary of what this dataset is. -->
+To benchmark our models, we utilized several publicaly available datasets, encompassing diverse protein-ligand interactions and binding affinity values. Key datasets include BindingDB (1D data with protein sequnces and SMILES), LP-PDBBind (containing 3D complexes), and other target-specific datasets such as USP7, MPro, and three targets from the protein-ligand free energy benchmark (SYK, HIF2A, and MCL1). These datasets capture a wide range of binding affinity measurements, allowing us to evaluate and compare model performance against traditional docking and free energy methods. All datasets have been meticulously cleaned and are available on Hugging Face as `BALM-Benchmark`.
+
+### BindingDB
+BindingDB provides experimental binding affinity data (Kd values) for protein-ligand interactions. We focused on $K_d$ values due to inconsistencies in other affinity types. After filtering for computational efficiency and data consistency, the dataset comprises around 25,000 interactions with ~1,070 unique targets and 9,200 ligands. We implemented four data splits (Random, Cold Target, Cold Drug, and Scaffold) to evaluate generalizability, with splits based on ligand or protein novelty and structural dissimilarity, guided by the Murcko scaffold approach.
+
+### LP-PDBBind
+Derived from PDBBind v2020, LP-PDBBind is a curated collection of ~20,000 protein-ligand structures with experimental binding data. This dataset was reorganized to reduce similarity across splits and cleaned to remove covalently bound ligands and rare atomic elements. To ensure model reliability, we used Clean Level 1 (CL1) for training and the higher-quality CL2 data for validation and testing as recomended [here](https://pubmed.ncbi.nlm.nih.gov/37645037/). Here we provide 1D data, for 3D complexes please download from [here](https://github.com/THGLab/LP-PDBBind/).
+
+### USP7
+The USP7 dataset, developed by [Shen et al.](https://jcheminf.biomedcentral.com/articles/10.1186/s13321-022-00675-8), contains binding data for USP7 inhibitors from ChEMBL. After processing to remove assay limits, it includes 1,799 ligands with experimentally measured affinities, provided as IC50 values and converted to pIC50 for consistency.
+
+### MPro
+Collected as part of the [COVID Moonshot project](https://www.science.org/doi/10.1126/science.abo7201), the MPro dataset focuses on inhibitors targeting the SARS-CoV-2 main protease. The final cleaned dataset includes 2,062 ligands with IC50 values, converted to pIC50 for stability in training.
+
+### Protein-Ligand Free Energy Benchmark
+Selected from the protein-ligand free energy benchmark by [Hahn et al.](https://livecomsjournal.org/index.php/livecoms/article/view/v4i1e1497), this dataset includes three targets: MCL1, HIF2A, and SYK. These targets offer diverse interactions, allowing for robust comparison with alchemical free energy methods. The datasets contain 37, 25, and 43 ligands, respectively, for benchmarking model predictions against established free energy methods.
+
 
 - **Dataset Repository:** https://huggingface.co/datasets/BALM/BALM-benchmark
 - **Code Repository:** https://github.com/meyresearch/BALM
@@ -220,73 +205,50 @@ BALM-Benchmark is a comprehensive benchmark suite that combines multiple seminal
     - **Drug** (`string`): Ligand sequence (i.e., SMILES string).
     - **Target_ID** (`string`): Index of the target protein from the TDC.
     - **Target** (`string`): Protein sequence (i.e., sequence of amino acids).
-    - **Y** (`float32`): binding affinity value in pKd.
-- **CATS**: 
-    - **Index** (`string`): Index of the ligand-target pair. 
-    - **Drug** (`string`): Ligand sequence (i.e., SMILES string).
-    - **Target** (`string`): Protein sequence (i.e., sequence of amino acids).
-    - **Y** (`float32`): binding affinity value in pKd.
-- **HIF2A**:
-    - **Index** (`string`): Index of the ligand-target pair. 
-    - **Y** (`float32`): binding affinity value in pKd.
-    - **Drug** (`string`): Ligand sequence (i.e., SMILES string).
-    - **Target** (`string`): Protein sequence (i.e., sequence of amino acids).
-- **HSP90**:
-    - **Index** (`string`): Index of the ligand-target pair. 
-    - **Drug** (`string`): Ligand sequence (i.e., SMILES string).
-    - **IC50 (nM)** (`float32`): binding affinity value in IC50. 
-    - **Target** (`string`): Protein sequence (i.e., sequence of amino acids).
-    - **Y** (`float32`): binding affinity value in pKd.
-- **LeakyPDB**:
-    - **Index** (`string`): Index of the ligand-target pair. 
-    - **header** (`string`): TBA 
-    - **smiles** (`string`): TBA 
-    - **category** (`string`): TBA 
-    - **seq** (`string`): TBA 
-    - **resolution** (`float32`): TBA 
-    - **date** (`string`): TBA 
-    - **type** (`string`): TBA 
-    - **new_split** (`string`): TBA 
-    - **CL1** (`bool`): TBA 
-    - **CL2** (`bool`): TBA 
-    - **CL3** (`bool`): TBA 
-    - **remove_for_balancing_val** (`bool`): TBA 
-    - **kd/ki** (`string`): TBA 
-    - **value** (`float32`): TBA 
-    - **covalent** (`bool`): TBA 
-- **MCL1**:
-    - **Index** (`string`): Index of the ligand-target pair. 
-    - **Y** (`float32`): binding affinity value in pKd.
-    - **Drug** (`string`): Ligand sequence (i.e., SMILES string).
-    - **Target** (`string`): Protein sequence (i.e., sequence of amino acids).
+    - **Y** (`float32`): binding affinity value in $pK_d$.
 - **Mpro**:
     - **Index** (`string`): Index of the ligand-target pair. 
-    - **Y** (`float32`): binding affinity value in pKd.
+    - **Y** (`float32`): binding affinity value in pIC50.
     - **Drug** (`string`): Ligand sequence (i.e., SMILES string).
     - **Target** (`string`): Protein sequence (i.e., sequence of amino acids).
-- **SYK**:
+- **USP7**:
     - **Index** (`string`): Index of the ligand-target pair. 
-    - **Y** (`float32`): binding affinity value in pKd.
+    - **Y** (`float32`): binding affinity value in pIC50.
     - **Drug** (`string`): Ligand sequence (i.e., SMILES string).
     - **Target** (`string`): Protein sequence (i.e., sequence of amino acids).
-- **USP7**:
+- **LeakyPDB**:
+    - **Index** (`string`): Identifier for each ligand-target pair in the dataset.
+    - **pdb_id** (`string`): Unique identifier for the protein structure in the Protein Data Bank (PDB).
+    - **Drug** (`string`): SMILES string representing the ligand's chemical structure.
+    - **category** (`string`): Classification category for the ligand-protein complex.
+    - **Target** (`string`): Protein sequence, represented as a sequence of amino acids.
+    - **resolution** (`float32`): Structural resolution of the protein-ligand complex, typically measured in angstroms.
+    - **date** (`string`): Date of structural determination or deposition in the PDB.
+    - **type** (`string`): Type or family classification of the protein.
+    - **new_split** (`string`): Specifies the split category for the LP-PDBBind dataset.
+    - **CL1** (`bool`): Boolean indicating whether the complex belongs to Clean Level 1 (CL1) in the LP-PDBBind dataset.
+    - **CL2** (`bool`): Boolean indicating whether the complex belongs to Clean Level 2 (CL2) in the LP-PDBBind dataset.
+    - **CL3** (`bool`): Boolean indicating whether the complex belongs to Clean Level 3 (CL3) in the LP-PDBBind dataset.
+    - **remove_for_balancing_val** (`bool`): Boolean indicating if the entry is excluded for balancing in validation sets.
+    - **kd/ki** (`string`): Original binding affinity measurement (Kd or Ki), typically with units.
+    - **Y** (`float32`): Binding affinity value provided in log scale ($pK_d$).
+    - **covalent** (`bool`): Boolean indicating if the ligand is covalently bound to the protein.
+- **HIF2A, MCL1, and SYK**:
     - **Index** (`string`): Index of the ligand-target pair. 
-    - **Y** (`float32`): binding affinity value in pKd.
+    - **Y** (`float32`): binding affinity value in pKi (for MCL1) and pIC50 (for HIF2A, and SYK).
     - **Drug** (`string`): Ligand sequence (i.e., SMILES string).
     - **Target** (`string`): Protein sequence (i.e., sequence of amino acids).
 
 
+
 ### Dataset Sources
 
-- **BindingDB_filtered**:
-- **CATS**: 
-- **HIF2A**:
-- **HSP90**:
-- **LeakyPDB**:
-- **MCL1**:
-- **Mpro**:
-- **SYK**:
-- **USP7**:
+- **BindingDB_filtered**: Derived from [Therapeutics Data Commons (TDC)](https://tdcommons.ai/), with additional filtering and cleaning to enhance consistency and computational efficiency.
+- **LeakyPDB**: Collected from the [LP-PDBBind repository](https://github.com/THGLab/LP-PDBBind/) and described in [this publication](https://pubmed.ncbi.nlm.nih.gov/37645037/).
+- **HIF2A, MCL1, and SYK**: Sourced from the protein-ligand benchmark dataset available on [GitHub](https://github.com/openforcefield/protein-ligand-benchmark) and detailed in the [LiveCoMS journal](https://livecomsjournal.org/index.php/livecoms/article/view/v4i1e1497).
+- **Mpro**: Data for SARS-CoV-2 main protease (Mpro) inhibitors sourced from [Science](https://www.science.org/doi/10.1126/science.abo7201).
+- **USP7**: Collected from ChEMBL and curated as described in this [Journal of Cheminformatics article](https://jcheminf.biomedcentral.com/articles/10.1186/s13321-022-00675-8).
+
 
 ## Uses
 
@@ -300,7 +262,7 @@ from datasets import load_dataset
 syk_data = load_dataset("BALM/BALM-benchmark", "SYK", split="train")
 ```
 
-As mentioned in the [Dataset Sources](#dataset-sources), the available datasets are:
+<!-- As mentioned in the [Dataset Sources](#dataset-sources), the available datasets are:
 
 - `BindingDB_filtered`
 - `CATS`
@@ -310,7 +272,7 @@ As mentioned in the [Dataset Sources](#dataset-sources), the available datasets
 - `MCL1`
 - `Mpro`
 - `SYK`
-- `USP7`
+- `USP7` -->
 
 Notice that all datasets only have one split (`train`). This is intentional such that the users can define their own splits, and can experiment with more random seeds for robustness.
 We highly recommend checking out different strategies for splitting the data (e.g., BindingDB) in [our BALM code repository](https://github.com/meyresearch/BALM/blob/refactor/balm/datasets/bindingdb_filtered.py#L157-L169).