esm/cookbook/tutorials/2_embed.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# [Tutorial 2](https://github.com/evolutionaryscale/esm/tree/main/cookbook/tutorials): Embedding with ESM C\n",
    "\n",
    "In this notebook we will see how to embed a batch of sequences using ESM C, as well as explore its different layers"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Install esm and other dependencies\n",
    "! pip install esm\n",
    "! pip install matplotlib"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Set up Forge client for ESM C"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Grab a token from [the Forge console](https://forge.evolutionaryscale.ai/console) and add it below. Note that your token is like a password for your account and you should take care to protect it. For this reason it is recommended to frequently create a new token and delete old, unused ones. It is also recommended to paste the token directly into an environment variable or use a utility like `getpass` as shown below so tokens are not accidentally shared or checked into code repositories."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from getpass import getpass\n",
    "\n",
    "token = getpass(\"Token from Forge console: \")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from esm.sdk import client\n",
    "\n",
    "model = client(\n",
    "    model=\"esmc-300m-2024-12\", url=\"https://forge.evolutionaryscale.ai\", token=token\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Set up utilities for embedding sequences\n",
    "\n",
    "Since we're embedding more than a few sequences, we're going to use a threaded async call to Forge and let Forge take care of batching and parallelization on the backend."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from concurrent.futures import ThreadPoolExecutor\n",
    "from typing import Sequence\n",
    "\n",
    "from esm.sdk.api import (\n",
    "    ESM3InferenceClient,\n",
    "    ESMProtein,\n",
    "    ESMProteinError,\n",
    "    LogitsConfig,\n",
    "    LogitsOutput,\n",
    "    ProteinType,\n",
    ")\n",
    "\n",
    "EMBEDDING_CONFIG = LogitsConfig(\n",
    "    sequence=True, return_embeddings=True, return_hidden_states=True\n",
    ")\n",
    "\n",
    "\n",
    "def embed_sequence(model: ESM3InferenceClient, sequence: str) -> LogitsOutput:\n",
    "    protein = ESMProtein(sequence=sequence)\n",
    "    protein_tensor = model.encode(protein)\n",
    "    output = model.logits(protein_tensor, EMBEDDING_CONFIG)\n",
    "    return output\n",
    "\n",
    "\n",
    "def batch_embed(\n",
    "    model: ESM3InferenceClient, inputs: Sequence[ProteinType]\n",
    ") -> Sequence[LogitsOutput]:\n",
    "    \"\"\"Forge supports auto-batching. So batch_embed() is as simple as running a collection\n",
    "    of embed calls in parallel using asyncio.\n",
    "    \"\"\"\n",
    "    with ThreadPoolExecutor() as executor:\n",
    "        futures = [\n",
    "            executor.submit(embed_sequence, model, protein) for protein in inputs\n",
    "        ]\n",
    "        results = []\n",
    "        for future in futures:\n",
    "            try:\n",
    "                results.append(future.result())\n",
    "            except Exception as e:\n",
    "                results.append(ESMProteinError(500, str(e)))\n",
    "    return results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Requesting a specific hidden layer\n",
    "\n",
    "ESM C 6B's hidden states are really large, so we only allow one specific layer to be requested per API call. This also works for other ESM C models, but it is required for ESM C 6B. \n",
    "Refer to https://forge.evolutionaryscale.ai/console to find the number of hidden layers for each model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ESMC_6B_EMBEDDING_CONFIG = LogitsConfig(return_hidden_states=True, ith_hidden_layer=55)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load dataset\n",
    "\n",
    "This dataset is taken from Muir, et al. 2024 [\"Evolutionary-Scale Enzymology Enables Biochemical Constant Prediction Across a Multi-Peaked Catalytic Landscape\"](https://doi.org/10.1101/2024.10.23.619915) which explores a model enzyme called Adenylate Kinase (ADK). Adenylate Kinase appears in many different organisms with different structural classes (referred to as its \"lid type\"). We'll embed this set of ADK sequences and see if we can recover known structural classes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--2025-01-16 00:34:10--  https://docs.google.com/uc?export=download&id=1SpOkL11MJxIgy99dqufvUNJuCiuhxuyg\n",
      "Resolving docs.google.com (docs.google.com)... 142.251.46.238, 2607:f8b0:4005:811::200e\n",
      "Connecting to docs.google.com (docs.google.com)|142.251.46.238|:443... connected.\n",
      "HTTP request sent, awaiting response... 303 See Other\n",
      "Location: https://drive.usercontent.google.com/download?id=1SpOkL11MJxIgy99dqufvUNJuCiuhxuyg&export=download [following]\n",
      "--2025-01-16 00:34:10--  https://drive.usercontent.google.com/download?id=1SpOkL11MJxIgy99dqufvUNJuCiuhxuyg&export=download\n",
      "Resolving drive.usercontent.google.com (drive.usercontent.google.com)... 142.250.191.33, 2607:f8b0:4005:80f::2001\n",
      "Connecting to drive.usercontent.google.com (drive.usercontent.google.com)|142.250.191.33|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 43132 (42K) [application/octet-stream]\n",
      "Saving to: ‘adk.csv’\n",
      "\n",
      "adk.csv             100%[===================>]  42.12K  --.-KB/s    in 0.02s   \n",
      "\n",
      "2025-01-16 00:34:13 (1.93 MB/s) - ‘adk.csv’ saved [43132/43132]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "!wget --no-check-certificate \"https://docs.google.com/uc?export=download&id=1SpOkL11MJxIgy99dqufvUNJuCiuhxuyg\" -O adk.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "adk_path = \"adk.csv\"\n",
    "df = pd.read_csv(adk_path)\n",
    "df = df[[\"org_name\", \"sequence\", \"lid_type\", \"temperature\"]]\n",
    "df = df[df[\"lid_type\"] != \"other\"]  # drop one structural class for simplicity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Retrying... Attempt 1 after 1.0s due to: (502, 'Failure in logits: {\"status\":\"error\",\"message\":\"Model unavailable - please retry\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (502, 'Failure in logits: {\"status\":\"error\",\"message\":\"Model unavailable - please retry\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (502, 'Failure in encode: {\"status\":\"error\",\"message\":\"Model unavailable - please retry\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 3 after 4.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 3 after 4.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 3 after 4.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 3 after 4.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 52 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 52 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 54 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 54 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 52 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 52 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 54 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 54 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in logits: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 52 requests for esmc-300m-2024-12 and logits.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 54 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 54 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 51 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 51 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 51 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 51 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 51 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (502, 'Failure in logits: {\"status\":\"error\",\"message\":\"Model unavailable - please retry\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 2 after 2.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (429, 'Failure in encode: {\"status\":\"error\",\"message\":\"You have exceeded your usage cap of 50 requests during the last minute with 50 requests for esmc-300m-2024-12 and encode.\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (502, 'Failure in logits: {\"status\":\"error\",\"message\":\"Model unavailable - please retry\"}')\n",
      "Retrying... Attempt 1 after 1.0s due to: (502, 'Failure in logits: {\"status\":\"error\",\"message\":\"Model unavailable - please retry\"}')Retrying... Attempt 1 after 1.0s due to: (502, 'Failure in encode: {\"status\":\"error\",\"message\":\"Model unavailable - please retry\"}')\n",
      "\n",
      "Retrying... Attempt 2 after 2.0s due to: (502, 'Failure in encode: {\"status\":\"error\",\"message\":\"Model unavailable - please retry\"}')\n"
     ]
    }
   ],
   "source": [
    "# You may see some error messages due to rate limits on each Forge account,\n",
    "# but this will retry until the embedding job is complete\n",
    "# This may take a few minutes to run\n",
    "outputs = batch_embed(model, df[\"sequence\"].tolist())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "embedding shape [num_layers, hidden_size]: torch.Size([31, 960])\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "\n",
    "# we'll summarize the embeddings using their mean across the sequence dimension\n",
    "# which allows us to compare embeddings for sequences of different lengths\n",
    "all_mean_embeddings = [\n",
    "    torch.mean(output.hidden_states, dim=-2).squeeze() for output in outputs\n",
    "]\n",
    "\n",
    "# now we have a list of tensors of [num_layers, hidden_size]\n",
    "print(\"embedding shape [num_layers, hidden_size]:\", all_mean_embeddings[0].shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Examine the performance of different layer embeddings\n",
    "\n",
    "For this example, we're going to use PCA to visualize whether the embeddings separate our proteins by their structural class. To assess the quality of our PCA, we fit a K means classifier with three clusters, corresponding to the three structural classes of our enzyme, and compute the [rand index](https://en.wikipedia.org/wiki/Rand_index), a measure of the quality of the clustering."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.metrics import adjusted_rand_score\n",
    "\n",
    "N_KMEANS_CLUSTERS = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_embeddings_at_layer(all_mean_embeddings: torch.Tensor, layer_idx: int):\n",
    "    stacked_mean_embeddings = torch.stack(\n",
    "        [embedding[layer_idx, :] for embedding in all_mean_embeddings]\n",
    "    ).numpy()\n",
    "\n",
    "    # project all the embeddings to 2D using PCA\n",
    "    pca = PCA(n_components=2)\n",
    "    pca.fit(stacked_mean_embeddings)\n",
    "    projected_mean_embeddings = pca.transform(stacked_mean_embeddings)\n",
    "\n",
    "    # compute kmeans purity as a measure of how good the clustering is\n",
    "    kmeans = KMeans(n_clusters=N_KMEANS_CLUSTERS, random_state=0).fit(\n",
    "        projected_mean_embeddings\n",
    "    )\n",
    "    rand_index = adjusted_rand_score(df[\"lid_type\"], kmeans.labels_)\n",
    "\n",
    "    # plot the clusters\n",
    "    plt.figure(figsize=(4, 4))\n",
    "    sns.scatterplot(\n",
    "        x=projected_mean_embeddings[:, 0],\n",
    "        y=projected_mean_embeddings[:, 1],\n",
    "        hue=df[\"lid_type\"],\n",
    "    )\n",
    "    plt.title(\n",
    "        f\"PCA of mean embeddings at layer {layer_idx}.\\nRand index: {rand_index:.2f}\"\n",
    "    )\n",
    "    plt.xlabel(\"PC 1\")\n",
    "    plt.ylabel(\"PC 2\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_embeddings_at_layer(all_mean_embeddings, layer_idx=30)\n",
    "plot_embeddings_at_layer(all_mean_embeddings, layer_idx=12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that the top principal components of layer 12 separate structural classes better than that of layer 30. Embed away! And keep in mind that different layers may be better or worse for your particular use-case."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "openesm",
   "language": "python",
   "name": "esmopen"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}