In our first step, we are working to train models that we can use for predicting good candidates and possible off-target effects. We are using the ASO Atlas developed by MRC CoRE researchers, to train a large language model (OligoAI) on preclinical drug screening data for ASOs.
