Update README.md

README.md

@@ -19,7 +19,7 @@ pipeline_tag: feature-extraction
 
 **GeneJEPA** is a Joint-Embedding Predictive Architecture (JEPA) trained for self-supervised representation learning on scRNA-seq. It uses a Perceiver-style encoder to handle sparse, high-dimensional gene count vectors and a Fourier-feature tokenizer for numerical tokenization.
 
-**Why?** Produce compact cell embeddings you can use for clustering, transfer learning, linear probes, and downstream biological tasks.
+**Why?** Produce compact cell embeddings you can use for clustering, transfer learning, linear probes, perturbation prediction, and downstream biological tasks.
 
 ---