Update README.md

README.md

@@ -29,7 +29,10 @@ tags:
 This model is trained to predict general binding sites of proteins using on the sequence. This is a finetuned version of 
 `esm2_t6_8M_UR50D`, trained on [this dataset](https://huggingface.co/datasets/AmelieSchreiber/general_binding_sites). The data is 
 not filtered by family, and thus the model may be overfit to some degree. In the Hugging Face Inference API widget to the right 
-there are three protein sequence examples. The first is a DNA binding protein, the second and third were obtained using [EvoProtGrad](https://github.com/Amelie-Schreiber/sampling_protein_language_models/blob/main/EvoProtGrad_copy.ipynb)
+there are three protein sequence examples. The first is a DNA binding protein ([see UniProt entry here](https://www.uniprot.org/uniprotkb/D3ZG52/entry)). 
+Note there is significant overlap in the predicted binding sites and the binding sites given in UniProt. 
+
+The second and third were obtained using [EvoProtGrad](https://github.com/Amelie-Schreiber/sampling_protein_language_models/blob/main/EvoProtGrad_copy.ipynb)
 a Markov Chain Monte Carlo method of (in silico) directed evolution of proteins based on a form of Gibbs sampling. The mutatant-type 
 protein sequences in theory should have similar binding sites to the wild-type protein sequence, but perhaps with higher binding affinity. 
 Testing this out on the model, we see the two proteins indeed have the same binding sites, which validates to some degree that the model