switch out biopython for bioservices due to backend apis it has access to (kegg, etc.)

app.py

@@ -51,6 +51,10 @@ class GradioUIWithExamples(GradioUI):
             gr.Markdown("## Biomedical Answers")
             gr.Markdown("""
             An experimental knowledge platform that attempts to transforms fragmented biomedical data into answers.
+
+            A word of caution... this is a journey of discovery to understand what the Smolagents coding approach can provide with a simple app and a handful of dependencies.  In terms of the responses provided it has a ways to go until I would claim that it provides true knowledge or insight.
+
+            If anything I think that it highlights the value of ontologies and associated vocabularies in guiding LLM behavior.  Probably a good time to go back and review some notes on DSPy and layer on domain vocabularies...
             """)
 
             input_box = gr.Textbox(