Update README.md

README.md

@@ -10,6 +10,10 @@ datasets:
 
 ---
 
+## Code:
+Code is made public at https://github.com/jsjung00/diffusePerturb. 
+A more detailed write up is also linked there.
+
 ### Overview and motivation  
 Gene expression count vectors are tricky to navigate with machine learning models: they have a very peaked distribution around small values but it is also long tailed. 
 Moreover, the count distributions can be cell dependent and higher variable per sample. This makes it challenging to train a model to predict count vectors. When predicting raw counts and moreover predicting the mean (say under L2 loss), without hand crafted loss function, the model will be dominated by the large counts and will not be robust to errors of small counts.