simona-rc commited on
Commit
843c231
1 Parent(s): 1937fef

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +74 -4
README.md CHANGED
@@ -1,8 +1,78 @@
1
  ---
2
  tags:
3
- - model_hub_mixin
 
 
 
 
 
 
 
 
 
 
 
 
 
4
  ---
 
 
 
 
5
 
6
- This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
7
- - Library: [More Information Needed]
8
- - Docs: [More Information Needed]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  tags:
3
+ - drug-discovery
4
+ - ibm
5
+ - mammal
6
+ - pytorch
7
+ - small molecules drugs
8
+ - smiles
9
+ - MoleculeNet
10
+ - FDA approval
11
+ - safetensors
12
+ - biomed-multi-alignment
13
+ license: apache-2.0
14
+ library_name: biomed-multi-alignment
15
+ base_model:
16
+ - ibm/biomed.omics.bl.sm.ma-ted-458m
17
  ---
18
+ Drugs must satisfy stringent criteria for both efficacy and safety. This model predicts the likelihood of FDA approval
19
+ for small-molecule drugs, represented using SMILES (Simplified Molecular Input Line Entry System) strings. It is a fine-tuned version of the
20
+ IBM biomedical foundation model, ibm/biomed.omics.bl.sm.ma-ted-458m [1], trained on over 2 billion biological samples spanning multiple modalities,
21
+ including proteins, small molecules, and single-cell gene expression data.
22
 
23
+ The fine-tuning was performed using the MoleculeNet Clintox dataset [2]. For benchmarking, we employed predefined training, validation, and testing splits
24
+ provided by MolFormer [3], sourced from the dataset referenced in [4].
25
+
26
+ - [1] https://huggingface.co/ibm/biomed.omics.bl.sm.ma-ted-458m
27
+ - [2] Zhenqin Wu et al. “MoleculeNet: a benchmark for molecular machine learning”.
28
+ In: Chemical science 9.2 (2018), pp. 513–530.
29
+ - [3] Jerret Ross et al. “Large-scale chemical language representations capture molecular
30
+ structure and properties”. In: Nature Machine Intelligence 4.12 (2022),
31
+ pp. 1256–1264.
32
+ - [4] https://github.com/IBM/molformer/tree/main/data that points to https://ibm.ent.box.com/v/MoLFormer-data (file: finetune datasets.zip).
33
+
34
+ ## Model Summary
35
+
36
+ - **Developers:** IBM Research
37
+ - **GitHub Repository:** https://github.com/BiomedSciAI/biomed-multi-alignment
38
+ - **Paper:** https://arxiv.org/abs/2410.22367
39
+ - **Release Date**: Dec 4th, 2024
40
+ - **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0).
41
+
42
+ ## Usage
43
+
44
+ Using `biomed.omics.bl.sm.ma-ted-458m.moleculenet_clintox_fda` requires installing https://github.com/BiomedSciAI/biomed-multi-alignment
45
+
46
+ ```
47
+ pip install git+https://github.com/BiomedSciAI/biomed-multi-alignment.git
48
+ ```
49
+
50
+ A simple example for using `ibm/omics.bl.sm.ma-ted-458m.moleculenet_clintox_fda`:
51
+
52
+ ```
53
+ from mammal.examples.molnet.molnet_infer import load_model, task_infer
54
+
55
+ smiles_seq = "C(Cl)Cl"
56
+
57
+ task_dict = load_model(task_name="FDA_APPR", device="cpu")
58
+ result = task_infer(task_dict=task_dict, smiles_seq=smiles_seq)
59
+ print(f"The prediction for {smiles_seq=} is {result}")
60
+ ```
61
+
62
+ See our detailed example at: on `https://github.com/BiomedSciAI/biomed-multi-alignment`
63
+
64
+
65
+ ## Citation
66
+
67
+ If you found our work useful, please consider giving a star to the repo and cite our paper:
68
+ ```
69
+ @misc{shoshan2024mammalmolecularaligned,
70
+ title={MAMMAL -- Molecular Aligned Multi-Modal Architecture and Language},
71
+ author={Yoel Shoshan and Moshiko Raboh and Michal Ozery-Flato and Vadim Ratner and Alex Golts and Jeffrey K. Weber and Ella Barkan and Simona Rabinovici-Cohen and Sagi Polaczek and Ido Amos and Ben Shapira and Liam Hazan and Matan Ninio and Sivan Ravid and Michael M. Danziger and Joseph A. Morrone and Parthasarathy Suryanarayanan and Michal Rosen-Zvi and Efrat Hexter},
72
+ year={2024},
73
+ eprint={2410.22367},
74
+ archivePrefix={arXiv},
75
+ primaryClass={q-bio.QM},
76
+ url={https://arxiv.org/abs/2410.22367},
77
+ }
78
+ ```