Use / licensing of scTranslator model in research projects

[edg1983]

Hello,

I’m contacting you to understand better if there is any restriction on using the scTranslator model for research purposes. I've tried sending you an email directly, but your email server blocks my account. Thus, I am trying to reach you here.

We are a research group based at Human Technopole in Milan, Italy, working on various projects at the interface of genomics and single-cell biology.

In our research group, we are currently testing various models for predicting protein levels from gene expression at single-cell resolution and found your solution is among the best performing for our task.

Starting from the model weights and the code available on your GitHub repository, we are extending it to make it more user-friendly and integrate it into an automated workflow. We are also running various comparisons on internal datasets. If our tests confirm the model performs well and generalizes well across datasets, we see 2 possible developments:

• use it to augment existing AML cancer single-cell datasets to aid discovery of useful treatment markers
• collect publicly available CITE-seq data and create a library of fine-tuned models based on different CITE-seq panels and a Python package for automated prediction based on your scTranslator.

Apart from proper citation and credit, is there any restriction on using scTranslator that applies to our scenario? Or is the model free to be used and incorporated into additional tools for research purposes?

Kind regards, Edoardo Giacopuzzi

[edg1983]

@ElaineLIU-920 Can you assist me with this?