Tutorial

We’ll show here how to explain molecular property prediction tasks without access to the gradients or any properties of a molecule. To set-up this activity, we need a black box model. We’ll use something simple here – the model is classifier that says if a molecule has an alcohol (1) or not (0). Let’s implement this model first

from rdkit import Chem
from rdkit.Chem.Draw import IPythonConsole

# set-up rdkit drawing preferences
IPythonConsole.ipython_useSVG = True
IPythonConsole.drawOptions.drawMolsSameScale = False


def model(smiles):
    mol = Chem.MolFromSmiles(smiles)
    match = mol.GetSubstructMatches(Chem.MolFromSmarts("[O;!H0]"))
    return 1 if match else 0

Let’s now try it out on some molecules

smi = "CCCCCCO"
print("f(s)", model(smi))
Chem.MolFromSmiles(smi)

f(s) 1

smi = "OCCCCCCO"
print("f(s)", model(smi))
Chem.MolFromSmiles(smi)

f(s) 1

smi = "c1ccccc1"
print("f(s)", model(smi))
Chem.MolFromSmiles(smi)

f(s) 0

Counterfacutal explanations

Let’s now explain the model - pretending we don’t know how it works - using counterfactuals

import exmol

instance = "CCCCCCO"
space = exmol.sample_space(instance, model, batched=False)

cfs = exmol.cf_explain(space, 1)
exmol.plot_cf(cfs)

We can see that removing the alcohol is the smallest change to affect the prediction of this molecule. Let’s see the space and look at where these counterfactuals are.

exmol.plot_space(space, cfs)

Explain using substructures

Now we’ll try to explain our model using substructures.

exmol.lime_explain(space)
exmol.plot_descriptors(space)

SMARTS annotations for MACCS descriptors were created using SMARTSviewer (smartsview.zbh.uni-hamburg.de, Copyright: ZBH, Center for Bioinformatics Hamburg) developed by K. Schomburg et. al. (J. Chem. Inf. Model. 2010, 50, 9, 1529–1535)

This seems like a pretty clear explanation. Let’s take a look at using substructures that are present in the molecule

import skunk

exmol.lime_explain(space, descriptor_type="ECFP")
svg = exmol.plot_descriptors(space, return_svg=True)
skunk.display(svg)
svg = exmol.plot_utils.similarity_map_using_tstats(space[0], return_svg=True)
skunk.display(svg)

We can see that most of the model is explained from the presence of the alcohol group - as expected.

Text

We can prepare a natural language summary of these results using exmol:

exmol.lime_explain(space, descriptor_type="ECFP")
e = exmol.text_explain(space)
for ei in e:
    print(ei[0], end="")

Is there primary alcohol? Yes and this is positively correlated with property. This is very important for the property

To prepare the natural language summary, we need to convert to a prompt that a model like GPT-3 can parse. Insert the output below into a language model to get a summary.

Or you can pass it directly, by installing the langchain package and setting-up an openai key

print(exmol.text_explain_generate(e, property_name="active"))

---------------------------------------------------------------------------
OpenAIError                               Traceback (most recent call last)
Cell In[11], line 1
----> 1 print(exmol.text_explain_generate(e, property_name="active"))

File /opt/hostedtoolcache/Python/3.11.10/x64/lib/python3.11/site-packages/exmol/exmol.py:1444, in text_explain_generate(text_explanations, property_name, llm_model, single)
   1435 prompt = prompt_template.format(property=property_name, text=text)
   1437 messages = [
   1438     {
   1439         "role": "system",
   (...)
   1442     {"role": "user", "content": prompt},
   1443 ]
-> 1444 response = openai.chat.completions.create(
   1445     model=llm_model,
   1446     messages=messages,
   1447     temperature=0.05,
   1448 )
   1450 return response.choices[0].message.content

File /opt/hostedtoolcache/Python/3.11.10/x64/lib/python3.11/site-packages/openai/_utils/_proxy.py:20, in LazyProxy.__getattr__(self, attr)
     19 def __getattr__(self, attr: str) -> object:
---> 20     proxied = self.__get_proxied__()
     21     if isinstance(proxied, LazyProxy):
     22         return proxied  # pyright: ignore

File /opt/hostedtoolcache/Python/3.11.10/x64/lib/python3.11/site-packages/openai/_utils/_proxy.py:55, in LazyProxy.__get_proxied__(self)
     54 def __get_proxied__(self) -> T:
---> 55     return self.__load__()

File /opt/hostedtoolcache/Python/3.11.10/x64/lib/python3.11/site-packages/openai/_module_client.py:12, in ChatProxy.__load__(self)
     10 @override
     11 def __load__(self) -> resources.Chat:
---> 12     return _load_client().chat

File /opt/hostedtoolcache/Python/3.11.10/x64/lib/python3.11/site-packages/openai/__init__.py:327, in _load_client()
    311         _client = _AzureModuleClient(  # type: ignore
    312             api_version=api_version,
    313             azure_endpoint=azure_endpoint,
   (...)
    323             http_client=http_client,
    324         )
    325         return _client
--> 327     _client = _ModuleClient(
    328         api_key=api_key,
    329         organization=organization,
    330         project=project,
    331         base_url=base_url,
    332         timeout=timeout,
    333         max_retries=max_retries,
    334         default_headers=default_headers,
    335         default_query=default_query,
    336         http_client=http_client,
    337     )
    338     return _client
    340 return _client

File /opt/hostedtoolcache/Python/3.11.10/x64/lib/python3.11/site-packages/openai/_client.py:105, in OpenAI.__init__(self, api_key, organization, project, base_url, timeout, max_retries, default_headers, default_query, http_client, _strict_response_validation)
    103     api_key = os.environ.get("OPENAI_API_KEY")
    104 if api_key is None:
--> 105     raise OpenAIError(
    106         "The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable"
    107     )
    108 self.api_key = api_key
    110 if organization is None:

OpenAIError: The api_key client option must be set either by passing api_key to the client or by setting the OPENAI_API_KEY environment variable