Demo/notebook for common use case

Perturbation-Response-Prediction / PRnet

PRnet is a flexible and scalable perturbation-conditioned generative model predicting transcriptional responses to unseen complex perturbations at bulk and single-cell levels.

Apache License 2.0

23 stars 3 forks source link

Demo/notebook for common use case #5

Open olarerin-lab opened 4 weeks ago

olarerin-lab commented 4 weeks ago

Hello @nicole1q , I have a small molecule with known structure that unfortunately was not tested in the original LINCS1000 dataset. I want to predict the expression profile of this small molecule in the same cell lines used in LINCS1000. Can you possibly create a demo in the README or an attached notebook highlighting how one would do this with your program. Many thanks in advance!

nicole1q commented 3 weeks ago

To prepare your dataset for a new small molecule, please refer to the instructions in custom_data_preprocessing.ipynb. Once your dataset is ready, you can predict the perturbed expression profile by running test_demo.py, adjusting the split_key argument as needed for your analysis.

If you’re interested in other applications or discussing potential collaborations, please feel free to reach out at biozy@ict.ac.cn.

olarerin-lab commented 3 weeks ago

Ok, i will take a look at the notebook. Do you have any documentation for the arguments to your programs? In particular "split_key"? What are the values it can take and what do they mean?

OMIC-coding commented 2 weeks ago

Same question~