IC2Bert: masked gene expression pretraining and supervised fine tuning for robust immune checkpoint blockade (ICB) response prediction



Summary

IC2Bert proposes a novel deep learning framework for predicting patient response to Immune Checkpoint Blockade (ICB) therapy. It leverages a BERT-like architecture pre-trained on a large, unlabeled gene expression dataset using a masked gene expression objective, allowing the model to learn general gene expression patterns. Subsequently, the pre-trained model is fine-tuned on labeled ICB response data, adapting it for the specific prediction task. This approach, IC2Bert, demonstrates superior performance and robustness compared to existing methods, particularly in handling diverse and limited datasets, showcasing the benefits of self-supervised pre-training for personalized ICB treatment.

Read more…

Credits: Source

This post is part of “Science/Immunology News”, Follow for more…!!!