diff --git a/data/tcr_epitope_binding/example_processing_and_templates.ipynb b/data/tcr_epitope_binding/example_processing_and_templates.ipynb
new file mode 100644
index 000000000..59528d11f
--- /dev/null
+++ b/data/tcr_epitope_binding/example_processing_and_templates.ipynb
@@ -0,0 +1,1486 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "27c08f91-3fa0-4800-8f6a-96a96f665cad",
+   "metadata": {},
+   "source": [
+    "# TCR-Epitope binding, Weber et al."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "6ef172b9-aad2-47da-bf4c-844a2a07ee8c",
+   "metadata": {},
+   "source": [
+    "Original data repository: https://tdcommons.ai/multi_pred_tasks/tcrepitope/"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7d18c95d-2ec6-45e1-addc-54a890097b8e",
+   "metadata": {},
+   "source": [
+    "# Imports"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "cf59e3e9-8061-4022-9eae-e978311b4155",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import yaml\n",
+    "from tdc.multi_pred import TCREpitopeBinding "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a6751ff9-2e3e-4d01-8395-7a5ae0c200d7",
+   "metadata": {},
+   "source": [
+    "# Data processing"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a1169ad2-e4bb-41c6-9625-6d1644c44a5b",
+   "metadata": {},
+   "source": [
+    "## Download data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "7bb8eb5e-f513-40d2-a68c-7cda1a51ad31",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "fn_data_original = \"data_original.csv\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "b39a142e-ccbc-49d2-98b0-a5f9bde9fd27",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Found local copy...\n",
+      "Loading...\n",
+      "Done!\n"
+     ]
+    }
+   ],
+   "source": [
+    "data = TCREpitopeBinding(name = 'weber', path = './data')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "26d9f62a-07f5-4113-8161-d5dfcf0bfb71",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "data.get_data().to_csv(fn_data_original, index=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "43873fc3-20a8-487d-a7c5-33bd58414159",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "total 30664\r\n",
+      "drwxr-xr-x  3 cody  staff    96B Mar 28 01:21 \u001b[34mdata\u001b[m\u001b[m\r\n",
+      "-rw-r--r--  1 cody  staff    15M Mar 28 01:22 data_original.csv\r\n",
+      "-rw-r--r--  1 cody  staff    47K Mar 28 01:18 example_processing_and_templates.ipynb\r\n",
+      "-rw-r--r--@ 1 cody  staff   5.3K Mar 28 01:18 transform.py\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "!ls -lh"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d9cda29a-a133-4f0e-992b-e77c9070ee93",
+   "metadata": {},
+   "source": [
+    "## Load original data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "77f614e7-b133-40bc-8759-2d930e4c120e",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "epitope_aa,epitope_smi,tcr,tcr_full,label\r\n",
+      "FLKEKGGL,CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC(=O)[C@H](CCC(=O)O)NC(=O)[C@H](CCCCN)NC(=O)[C@H](CC(C)C)NC(=O)[C@@H](N)Cc1ccccc1)C(=O)O,CSVWGTGKTYEQYF,SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLIATANQGSEATYESGFVIDKFPISRPNLTFSTLTVSNMSPEDSSIYLCSVWGTGKTYEQYFGPGTRLTVT,1\r\n",
+      "FLKEKGGL,CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC(=O)[C@H](CCC(=O)O)NC(=O)[C@H](CCCCN)NC(=O)[C@H](CC(C)C)NC(=O)[C@@H](N)Cc1ccccc1)C(=O)O,CSVWGEGRSYEQYF,SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLIATANQGSEATYESGFVIDKFPISRPNLTFSTLTVSNMSPEDSSIYLCSVWGEGRSYEQYFGPGTRLTVT,1\r\n",
+      "FLKEKGGL,CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC(=O)[C@H](CCC(=O)O)NC(=O)[C@H](CCCCN)NC(=O)[C@H](CC(C)C)NC(=O)[C@@H](N)Cc1ccccc1)C(=O)O,CSATILAGVPYGEQYF,GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLMLMATSNEGSKATYEQGVEKDKFLINHASLTLSTLTVTSAHPEDSSFYICSATILAGVPYGEQYFGPGTRLTVT,1\r\n",
+      "FLKEKGGL,CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC(=O)[C@H](CCC(=O)O)NC(=O)[C@H](CCCCN)NC(=O)[C@H](CC(C)C)NC(=O)[C@@H](N)Cc1ccccc1)C(=O)O,CSASEGTSSYEQYF,GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLMLMATSNEGSKATYEQGVEKDKFLINHASLTLSTLTVTSAHPEDSSFYICSASEGTSSYEQYFGPGTRLTVT,1\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "!head -n 5 {fn_data_original}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "8f5a0387-f9e3-4e1a-8d14-5df618195f70",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "df = pd.read_csv(fn_data_original, delimiter=\",\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "55b0bd63-62a0-469e-9d8a-e9ada3fe01c4",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>epitope_aa</th>\n",
+       "      <th>epitope_smi</th>\n",
+       "      <th>tcr</th>\n",
+       "      <th>tcr_full</th>\n",
+       "      <th>label</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGTGKTYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGEGRSYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSATILAGVPYGEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSASEGTSSYEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CASSFDREVTGELFF</td>\n",
+       "      <td>GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  epitope_aa                                        epitope_smi  \\\n",
+       "0   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "1   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "2   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "3   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "4   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "\n",
+       "                tcr                                           tcr_full  label  \n",
+       "0    CSVWGTGKTYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...      1  \n",
+       "1    CSVWGEGRSYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...      1  \n",
+       "2  CSATILAGVPYGEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...      1  \n",
+       "3    CSASEGTSSYEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...      1  \n",
+       "4   CASSFDREVTGELFF  GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...      1  "
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "95158ac1-05d7-4a21-b8e4-7f720306d331",
+   "metadata": {},
+   "source": [
+    "## Add column = field names\n",
+    "Clean column names (`fields_clean`) and keep original names (`fields_orig`)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "ec2458e5-455f-4f03-8ce9-c0d12e9ed371",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['epitope_aa', 'epitope_smi', 'tcr', 'tcr_full', 'label']"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "fields_orig = df.columns.tolist()\n",
+    "fields_orig"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "a46dd8ff-37b3-4894-8226-3bf98226dd09",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "    fields_clean = [\n",
+    "        \"epitope_aa\",\n",
+    "        \"epitope_smiles\",\n",
+    "        \"tcr\",\n",
+    "        \"tcr_full\",\n",
+    "        \"binding\",\n",
+    "    ]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "785d37cb-1fb4-4a91-a923-d5a78a37f36a",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "df.columns = fields_clean"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "1bf212cb-1653-457b-9f5d-416d4dd14b53",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>epitope_aa</th>\n",
+       "      <th>epitope_smiles</th>\n",
+       "      <th>tcr</th>\n",
+       "      <th>tcr_full</th>\n",
+       "      <th>binding</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGTGKTYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGEGRSYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSATILAGVPYGEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSASEGTSSYEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CASSFDREVTGELFF</td>\n",
+       "      <td>GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  epitope_aa                                     epitope_smiles  \\\n",
+       "0   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "1   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "2   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "3   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "4   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "\n",
+       "                tcr                                           tcr_full  \\\n",
+       "0    CSVWGTGKTYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...   \n",
+       "1    CSVWGEGRSYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...   \n",
+       "2  CSATILAGVPYGEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...   \n",
+       "3    CSASEGTSSYEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...   \n",
+       "4   CASSFDREVTGELFF  GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...   \n",
+       "\n",
+       "   binding  \n",
+       "0        1  \n",
+       "1        1  \n",
+       "2        1  \n",
+       "3        1  \n",
+       "4        1  "
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "1bfaeb22-26fb-4964-a71f-cae8335e5372",
+   "metadata": {},
+   "source": [
+    "## Data cleaning"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "7e746003-cb1f-434f-bba6-00f0c439c4ac",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "df.epitope_aa = (\n",
+    "    df.epitope_aa.str.strip()\n",
+    ")  # remove leading and trailing white space characters"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "d544fa60-343e-40e1-bd0c-4750f07a7145",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "assert not df.duplicated().sum()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "bc6f52c1-e0f6-48b3-95f4-e36d9a5ecde8",
+   "metadata": {},
+   "source": [
+    "## Save to csv"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "d6d5efa5-b4b4-4a25-8626-e10f3d691e83",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "fn_data_csv = \"data_clean.csv\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "727f8d7b-cbb6-43c7-9eab-9d4d65be6b3f",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "df.to_csv(fn_data_csv, index=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "id": "63c8d4a4-906e-418d-be39-879365b4dfa0",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "-rw-r--r--  1 cody  staff    15M Mar 28 01:23 data_clean.csv\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "!ls -lh {fn_data_csv}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "a51b9001-25d7-4e0e-a607-477cfc4a9f1c",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "epitope_aa,epitope_smiles,tcr,tcr_full,binding\r\n",
+      "FLKEKGGL,CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC(=O)[C@H](CCC(=O)O)NC(=O)[C@H](CCCCN)NC(=O)[C@H](CC(C)C)NC(=O)[C@@H](N)Cc1ccccc1)C(=O)O,CSVWGTGKTYEQYF,SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLIATANQGSEATYESGFVIDKFPISRPNLTFSTLTVSNMSPEDSSIYLCSVWGTGKTYEQYFGPGTRLTVT,1\r\n",
+      "FLKEKGGL,CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC(=O)[C@H](CCC(=O)O)NC(=O)[C@H](CCCCN)NC(=O)[C@H](CC(C)C)NC(=O)[C@@H](N)Cc1ccccc1)C(=O)O,CSVWGEGRSYEQYF,SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLIATANQGSEATYESGFVIDKFPISRPNLTFSTLTVSNMSPEDSSIYLCSVWGEGRSYEQYFGPGTRLTVT,1\r\n",
+      "FLKEKGGL,CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC(=O)[C@H](CCC(=O)O)NC(=O)[C@H](CCCCN)NC(=O)[C@H](CC(C)C)NC(=O)[C@@H](N)Cc1ccccc1)C(=O)O,CSATILAGVPYGEQYF,GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLMLMATSNEGSKATYEQGVEKDKFLINHASLTLSTLTVTSAHPEDSSFYICSATILAGVPYGEQYFGPGTRLTVT,1\r\n",
+      "FLKEKGGL,CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC(=O)[C@H](CCC(=O)O)NC(=O)[C@H](CCCCN)NC(=O)[C@H](CC(C)C)NC(=O)[C@@H](N)Cc1ccccc1)C(=O)O,CSASEGTSSYEQYF,GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLMLMATSNEGSKATYEQGVEKDKFLINHASLTLSTLTVTSAHPEDSSFYICSASEGTSSYEQYFGPGTRLTVT,1\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "!head -n 5 {fn_data_csv}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "id": "1a512943-4909-4d56-867d-50c151d8d607",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>epitope_aa</th>\n",
+       "      <th>epitope_smiles</th>\n",
+       "      <th>tcr</th>\n",
+       "      <th>tcr_full</th>\n",
+       "      <th>binding</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGTGKTYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGEGRSYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSATILAGVPYGEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSASEGTSSYEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CASSFDREVTGELFF</td>\n",
+       "      <td>GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  epitope_aa                                     epitope_smiles  \\\n",
+       "0   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "1   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "2   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "3   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "4   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "\n",
+       "                tcr                                           tcr_full  \\\n",
+       "0    CSVWGTGKTYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...   \n",
+       "1    CSVWGEGRSYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...   \n",
+       "2  CSATILAGVPYGEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...   \n",
+       "3    CSASEGTSSYEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...   \n",
+       "4   CASSFDREVTGELFF  GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...   \n",
+       "\n",
+       "   binding  \n",
+       "0        1  \n",
+       "1        1  \n",
+       "2        1  \n",
+       "3        1  \n",
+       "4        1  "
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "f3d730ce-fce0-49df-9eb8-b917e945fa9a",
+   "metadata": {},
+   "source": [
+    "## Load from csv"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "id": "077b0c5f-8772-4879-9317-3fa28799689b",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "fn_data_csv = \"data_clean.csv\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "id": "6eaef0e6-2115-4793-ac43-a196b25d47a0",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "df = pd.read_csv(fn_data_csv)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "43619e7c-9c82-4ff0-ae25-403861304635",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>epitope_aa</th>\n",
+       "      <th>epitope_smiles</th>\n",
+       "      <th>tcr</th>\n",
+       "      <th>tcr_full</th>\n",
+       "      <th>binding</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGTGKTYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGEGRSYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSATILAGVPYGEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSASEGTSSYEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CASSFDREVTGELFF</td>\n",
+       "      <td>GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  epitope_aa                                     epitope_smiles  \\\n",
+       "0   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "1   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "2   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "3   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "4   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "\n",
+       "                tcr                                           tcr_full  \\\n",
+       "0    CSVWGTGKTYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...   \n",
+       "1    CSVWGEGRSYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...   \n",
+       "2  CSATILAGVPYGEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...   \n",
+       "3    CSASEGTSSYEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...   \n",
+       "4   CASSFDREVTGELFF  GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...   \n",
+       "\n",
+       "   binding  \n",
+       "0        1  \n",
+       "1        1  \n",
+       "2        1  \n",
+       "3        1  \n",
+       "4        1  "
+      ]
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2f06e57c-02c5-493b-af65-c8bb9ac59421",
+   "metadata": {},
+   "source": [
+    "# meta YAML"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "id": "49771077-471d-4d71-a9a7-d6b094bbc4f3",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>epitope_aa</th>\n",
+       "      <th>epitope_smiles</th>\n",
+       "      <th>tcr</th>\n",
+       "      <th>tcr_full</th>\n",
+       "      <th>binding</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGTGKTYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSVWGEGRSYEQYF</td>\n",
+       "      <td>SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSATILAGVPYGEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CSASEGTSSYEQYF</td>\n",
+       "      <td>GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>FLKEKGGL</td>\n",
+       "      <td>CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...</td>\n",
+       "      <td>CASSFDREVTGELFF</td>\n",
+       "      <td>GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  epitope_aa                                     epitope_smiles  \\\n",
+       "0   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "1   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "2   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "3   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "4   FLKEKGGL  CC(C)C[C@H](NC(=O)CNC(=O)CNC(=O)[C@H](CCCCN)NC...   \n",
+       "\n",
+       "                tcr                                           tcr_full  \\\n",
+       "0    CSVWGTGKTYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...   \n",
+       "1    CSVWGEGRSYEQYF  SAVISQKPSRDICQRGTSLTIQCQVDSQVTMMFWYRQQPGQSLTLI...   \n",
+       "2  CSATILAGVPYGEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...   \n",
+       "3    CSASEGTSSYEQYF  GAVVSQHPSWVICKSGTSVKIECRSLDFQATTMFWYRQFPKQSLML...   \n",
+       "4   CASSFDREVTGELFF  GAGVSQTPSNKVTEKGKYVELRCDPISGHTALYWYRQSLGQGPEFL...   \n",
+       "\n",
+       "   binding  \n",
+       "0        1  \n",
+       "1        1  \n",
+       "2        1  \n",
+       "3        1  \n",
+       "4        1  "
+      ]
+     },
+     "execution_count": 23,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "id": "d3890961-444e-4a26-b8fc-ed8c4e959af9",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "meta = {\n",
+    "    \"name\": \"tcr_epitope_binding\",  # unique identifier, we will also use this for directory names\n",
+    "    \"description\": \"\"\"T-cells are an integral part of the adaptive immune system, whose survival, proliferation, \n",
+    "    activationand function are all governed by the interaction of their T-cell receptor (TCR) with immunogenic \n",
+    "    peptides (epitopes).A large repertoire of T-cell receptors with different specificity is needed to provide \n",
+    "    protection against a wide range of pathogens.This new task aims to predict the binding affinity \n",
+    "    given a pair of TCR sequence and epitope sequence.\"\"\",\n",
+    "    \"targets\": [\n",
+    "        {\n",
+    "            \"id\": \"binding\",  # name of the column in a tabular dataset\n",
+    "            \"description\": \"TCR epitope binding.\",  # description of what this column means\n",
+    "            \"units\": \"\",  # units of the values in this column (leave empty if unitless)\n",
+    "            \"type\": \"binary classification\",  # can be \"categorical\", \"ordinal\", \"continuous\"\n",
+    "            \"names\": [  # names for the property (to sample from for building the prompts)\n",
+    "                \"tcr binding affinity\",\n",
+    "                \"binding affinity\",\n",
+    "                \"binding\",\n",
+    "                \"epitope binding affinity\",\n",
+    "                \"epitope binding\"\n",
+    "\n",
+    "            ],\n",
+    "        },\n",
+    "    ],\n",
+    "    \"identifiers\": [\n",
+    "        {\n",
+    "            \"id\": \"epitope_smiles\",  # column name\n",
+    "            \"type\": \"SMILES\",  # can be \"SMILES\", \"SELFIES\", \"IUPAC\", \"OTHER\"\n",
+    "            \"description\": \"epitope smiles\",  # description (optional, except for \"OTHER\")\n",
+    "        },\n",
+    "        {\n",
+    "            \"id\": \"epitope_aa\",\n",
+    "            \"type\": \"Other\",\n",
+    "            \"description\": \"epitope amino acid sequence\",\n",
+    "\n",
+    "        },\n",
+    "        {\n",
+    "            \"id\": \"tcr\",\n",
+    "            \"type\": \"Other\",\n",
+    "            \"description\": \"hypervariable CDR3 loop\",\n",
+    "\n",
+    "        },\n",
+    "        {\n",
+    "            \"id\": \"tcr_full\",\n",
+    "            \"type\": \"Other\",\n",
+    "            \"description\": \"tcr full amino acid sequence\",\n",
+    "\n",
+    "        },\n",
+    "    ],\n",
+    "    \"templates\": [\n",
+    "        {\n",
+    "            \"prompt\": \"Please answer the following question.\\nPredict the <expt_value_text> for <molecule>.\",\n",
+    "            \"completion\": \"<exp_value>\"\n",
+    "\n",
+    "        }\n",
+    "    ],\n",
+    "    \"fields\": [\n",
+    "        {\n",
+    "            \"exp_value\": [\n",
+    "             \"values\": [ \n",
+    "            {\n",
+    "                \"name\": \"exp_value\",\n",
+    "                \"column\": \"exp_value\",\n",
+    "                \"text\": \"binding\"\n",
+    "            }\n",
+    "            ]\n",
+    "            ],\n",
+    "            \"molecule\": [\n",
+    "            \"values\": [\n",
+    "            {\n",
+    "                \"name\": \"epitope_smiles\",\n",
+    "                \"column\": \"epitope_smiles\",\n",
+    "                \"text\": \"SMILES\"\n",
+    "\n",
+    "            },\n",
+    "            {\n",
+    "                \"name\": \"tcr_full\",\n",
+    "                \"column\": \"tcr_full\",\n",
+    "                \"text\":\n",
+    "\n",
+    "            },\n",
+    "\n",
+    "        ]\n",
+    "        ],\n",
+    "\n",
+    "        }\n",
+    "    ],\n",
+    "    \"license\": \"CC BY 4.0\",  # license under which the original dataset was published\n",
+    "    \"links\": [  # list of relevant links (original dataset, other uses, etc.)\n",
+    "        {\n",
+    "            \"url\": \"https://tdcommons.ai/multi_pred_tasks/tcrepitope/\",\n",
+    "            \"description\": \"original data set link\",\n",
+    "        },\n",
+    "        {\n",
+    "            \"url\": \"https://doi.org/10.1093/bioinformatics/btab294\",\n",
+    "            \"description\": \"corresponding publication\",\n",
+    "        },\n",
+    "    ],\n",
+    "    \"benchmarks\": [\n",
+    "        {\n",
+    "            \"name\": \"TDC\",\n",
+    "            \"link\": \"https://tdcommons.ai/\",\n",
+    "            \"split_column\": \"split\",\n",
+    "        },\n",
+    "    ],\n",
+    "    \"num_points\": len(df),  # number of datapoints in this dataset\n",
+    "    \"bibtex\": [\n",
+    "        \"\"\"@article{weber2021titan,\n",
+    "        title={TITAN: T-cell receptor specificity prediction with bimodal attention network},\n",
+    "        author={Weber Anna,Born Janis, Martinez Maria Rodriguez},\n",
+    "        journal={Bioinformatics},\n",
+    "        volume={56},\n",
+    "        number={4},\n",
+    "        pages={i237-i234},\n",
+    "        year={2021},\n",
+    "        publisher={Oxford Academic}\n",
+    "        }\"\"\",\n",
+    "    ],\n",
+    "}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "id": "ec455cf0-962a-4c0d-bb3e-066e415ffd9b",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "def str_presenter(dumper, data):\n",
+    "    \"\"\"configures yaml for dumping multiline strings\n",
+    "    Ref: https://stackoverflow.com/questions/8640959/how-can-i-control-what-scalar-form-pyyaml-uses-for-my-data\n",
+    "    \"\"\"\n",
+    "    if data.count(\"\\n\") > 0:  # check for multiline string\n",
+    "        return dumper.represent_scalar(\"tag:yaml.org,2002:str\", data, style=\"|\")\n",
+    "    return dumper.represent_scalar(\"tag:yaml.org,2002:str\", data)\n",
+    "\n",
+    "\n",
+    "yaml.add_representer(str, str_presenter)\n",
+    "yaml.representer.SafeRepresenter.add_representer(\n",
+    "    str, str_presenter\n",
+    ")  # to use with safe_dum"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "id": "580bbd79-4845-4515-be94-3e4a9815d048",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "fn_meta = \"meta.yaml\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "id": "873fa5dd-9b60-40f5-b537-4d7a206414ea",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "with open(fn_meta, \"w\") as f:\n",
+    "    yaml.dump(meta, f, sort_keys=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "id": "d01686c0-6746-4fc4-b019-350270dfc26f",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "-rw-r--r--  1 cody  staff   1.7K Mar 28 01:23 meta.yaml\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "!ls -lh {fn_meta}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "id": "ef6063c5-7a8b-4344-bccf-a073443feebf",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "name: tcr_epitope_binding\r\n",
+      "description: \"T-cells are an integral part of the adaptive immune system, whose survival,\\\r\n",
+      "  \\ proliferation, \\n    activationand function are all governed by the interaction\\\r\n",
+      "  \\ of their T-cell receptor (TCR) with immunogenic \\n    peptides (epitopes).A large\\\r\n",
+      "  \\ repertoire of T-cell receptors with different specificity is needed to provide\\\r\n",
+      "  \\ \\n    protection against a wide range of pathogens.This new task aims to predict\\\r\n",
+      "  \\ the binding affinity \\n    given a pair of TCR sequence and epitope sequence.\"\r\n",
+      "targets:\r\n",
+      "- id: binding\r\n",
+      "  description: TCR epitope binding.\r\n",
+      "  units: ''\r\n",
+      "  type: binary classification\r\n",
+      "  names:\r\n",
+      "  - tcr binding affinity\r\n",
+      "  - binding affinity\r\n",
+      "  - binding\r\n",
+      "  - epitope binding affinity\r\n",
+      "  - epitope binding\r\n",
+      "identifiers:\r\n",
+      "- id: epitope_smiles\r\n",
+      "  type: SMILES\r\n",
+      "  description: epitope smiles\r\n",
+      "- id: epitope_aa\r\n",
+      "  type: Other\r\n",
+      "  description: epitope amino acid sequence\r\n",
+      "- id: tcr\r\n",
+      "  type: Other\r\n",
+      "  description: hypervariable CDR3 loop\r\n",
+      "- id: tcr_full\r\n",
+      "  type: Other\r\n",
+      "  description: tcr full amino acid sequence\r\n",
+      "license: CC BY 4.0\r\n",
+      "links:\r\n",
+      "- url: https://tdcommons.ai/multi_pred_tasks/tcrepitope/\r\n",
+      "  description: original data set link\r\n",
+      "- url: https://doi.org/10.1093/bioinformatics/btab294\r\n",
+      "  description: corresponding publication\r\n",
+      "benchmarks:\r\n",
+      "- name: TDC\r\n",
+      "  link: https://tdcommons.ai/\r\n",
+      "  split_column: split\r\n",
+      "num_points: 47182\r\n",
+      "bibtex:\r\n",
+      "- |-\r\n",
+      "  @article{weber2021titan,\r\n",
+      "          title={TITAN: T-cell receptor specificity prediction with bimodal attention network},\r\n",
+      "          author={Weber Anna,Born Janis, Martinez Maria Rodriguez},\r\n",
+      "          journal={Bioinformatics},\r\n",
+      "          volume={56},\r\n",
+      "          number={4},\r\n",
+      "          pages={i237-i234},\r\n",
+      "          year={2021},\r\n",
+      "          publisher={Oxford Academic}\r\n",
+      "          }\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "!cat {fn_meta}"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "bd3f930a-638b-4bb7-a1d2-80688f2f6891",
+   "metadata": {},
+   "source": [
+    "# create transform.py"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "id": "9aab00fd-58a8-40b0-be30-1e269e0d323b",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "path_file = \"transform.py\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "id": "8368bb20-8e1c-4b7d-b0e2-b39da36b5972",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Overwriting transform.py\n"
+     ]
+    }
+   ],
+   "source": [
+    "%%writefile $path_file\n",
+    "import pandas as pd\n",
+    "import yaml\n",
+    "from tdc.multi_pred import TCREpitopeBinding \n",
+    "def get_and_transform_data():\n",
+    "    # get raw data\n",
+    "    data = TCREpitopeBinding(name = 'weber', path = './data')\n",
+    "    \n",
+    "    split = data.get_split()\n",
+    "    df_train=split['train']\n",
+    "    df_valid=split['valid']\n",
+    "    df_test=split['test']\n",
+    "    df_train['split']=\"train\"\n",
+    "    df_valid['split']=\"valid\"\n",
+    "    df_test['split']=\"test\"\n",
+    "    df=pd.concat([df_train,df_valid,df_test],axis=0)\n",
+    "\n",
+    "    # create dataframenot necessary but ensure we can load the saved data\n",
+    "\n",
+    "    # check if fields are the same\n",
+    "    fields_orig = df.columns.tolist()\n",
+    "    assert fields_orig == [\n",
+    "        \"epitope_aa\",\n",
+    "        \"epitope_smi\",\n",
+    "        \"tcr\",\n",
+    "        \"tcr_full\",\n",
+    "        \"label\",\n",
+    "        \"split\"\n",
+    "    ]\n",
+    "\n",
+    "    # overwrite column names = fields\n",
+    "    fields_clean = [\n",
+    "        \"epitope_aa\",\n",
+    "        \"epitope_smiles\",\n",
+    "        \"tcr\",\n",
+    "        \"tcr_full\",\n",
+    "        \"binding\",\n",
+    "        \"split\"\n",
+    "    ]\n",
+    "    df.columns = fields_clean\n",
+    "\n",
+    "    # data cleaning\n",
+    "    df.epitope_aa = (\n",
+    "        df.epitope_aa.str.strip()\n",
+    "    )  # remove leading and trailing white space characters\n",
+    "\n",
+    "    assert not df.duplicated().sum()\n",
+    "\n",
+    "    # save to csv\n",
+    "    fn_data_csv = \"data_clean.csv\"\n",
+    "    df.to_csv(fn_data_csv, index=False)\n",
+    "\n",
+    "    # create meta yaml\n",
+    "    meta = {\n",
+    "        \"name\": \"tcr_epitope_binding\",  # unique identifier, we will also use this for directory names\n",
+    "        \"description\": \"\"\"T-cells are an integral part of the adaptive immune system, whose survival, proliferation, activation\n",
+    "        and function are all governed by the interaction of their T-cell receptor (TCR) with immunogenic peptides (epitopes).\n",
+    "        A large repertoire of T-cell receptors with different specificity is needed to provide protection against a wide range of pathogens.\n",
+    "        This new task aims to predict the binding affinity given a pair of TCR sequence and epitope sequence.\"\"\",\n",
+    "        \"targets\": [\n",
+    "            {\n",
+    "                \"id\": \"binding\",  # name of the column in a tabular dataset\n",
+    "                \"description\": \"TCR epitope binding.\",  # description of what this column means\n",
+    "                \"units\": \"\",  # units of the values in this column (leave empty if unitless)\n",
+    "                \"type\": \"binary classification\",  # can be \"categorical\", \"ordinal\", \"continuous\"\n",
+    "                \"names\": [  # names for the property (to sample from for building the prompts)\n",
+    "                    \"tcr binding affinity\",\n",
+    "                    \"binding affinity\",\n",
+    "                    \"binding\",\n",
+    "                    \"epitope binding affinity\",\n",
+    "                    \"epitope binding\"\n",
+    "\n",
+    "                ],\n",
+    "            },\n",
+    "        ],\n",
+    "        \"identifiers\": [\n",
+    "            {\n",
+    "                \"id\": \"epitope_smiles\",  # column name\n",
+    "                \"type\": \"SMILES\",  # can be \"SMILES\", \"SELFIES\", \"IUPAC\", \"OTHER\"\n",
+    "                \"description\": \"epitope smiles\",  # description (optional, except for \"OTHER\")\n",
+    "            },\n",
+    "            {\n",
+    "                \"id\": \"epitope_aa\",\n",
+    "                \"type\": \"Other\",\n",
+    "                \"description\": \"epitope amino acid sequence\",\n",
+    "    \n",
+    "            },\n",
+    "            {\n",
+    "                \"id\": \"tcr\",\n",
+    "                \"type\": \"Other\",\n",
+    "                \"description\": \"hypervariable CDR3 loop\",\n",
+    "    \n",
+    "            },\n",
+    "            {\n",
+    "                \"id\": \"tcr_full\",\n",
+    "                \"type\": \"Other\",\n",
+    "                \"description\": \"tcr full amino acid sequence\",\n",
+    "    \n",
+    "            },\n",
+    "        ],\n",
+    "        \"license\": \"CC BY 4.0\",  # license under which the original dataset was published\n",
+    "        \"links\": [  # list of relevant links (original dataset, other uses, etc.)\n",
+    "            {\n",
+    "                \"url\": \"https://tdcommons.ai/multi_pred_tasks/tcrepitope/\",\n",
+    "                \"description\": \"original data set link\",\n",
+    "            },\n",
+    "            {\n",
+    "                \"url\": \"https://doi.org/10.1093/bioinformatics/btab294\",\n",
+    "                \"description\": \"corresponding publication\",\n",
+    "            },\n",
+    "        ],\n",
+    "        \"benchmarks\": [\n",
+    "            {\n",
+    "                \"name\": \"TDC\",\n",
+    "                \"link\": \"https://tdcommons.ai/\",\n",
+    "                \"split_column\": \"split\",\n",
+    "            },\n",
+    "        ],\n",
+    "        \"num_points\": len(df),  # number of datapoints in this dataset\n",
+    "        \"bibtex\": [\n",
+    "            \"\"\"@article{weber2021titan,\n",
+    "            title={TITAN: T-cell receptor specificity prediction with bimodal attention network},\n",
+    "            author={Weber Anna,Born Janis, Martinez Maria Rodriguez},\n",
+    "            journal={Bioinformatics},\n",
+    "            volume={56},\n",
+    "            number={4},\n",
+    "            pages={i237-i234},\n",
+    "            year={2021},\n",
+    "            publisher={Oxford Academic}\n",
+    "            }\"\"\",\n",
+    "        ],\n",
+    "    }\n",
+    "\n",
+    "    def str_presenter(dumper, data):\n",
+    "        \"\"\"configures yaml for dumping multiline strings\n",
+    "        Ref: https://stackoverflow.com/questions/8640959/how-can-i-control-what-scalar-form-pyyaml-uses-for-my-data\n",
+    "        \"\"\"\n",
+    "\n",
+    "        if data.count(\"\\n\") > 0:  # check for multiline string\n",
+    "            return dumper.represent_scalar(\"tag:yaml.org,2002:str\", data, style=\"|\")\n",
+    "        return dumper.represent_scalar(\"tag:yaml.org,2002:str\", data)\n",
+    "\n",
+    "    yaml.add_representer(str, str_presenter)\n",
+    "    yaml.representer.SafeRepresenter.add_representer(\n",
+    "        str, str_presenter\n",
+    "    )  # to use with safe_dum\n",
+    "    fn_meta = \"meta.yaml\"\n",
+    "    with open(fn_meta, \"w\") as f:\n",
+    "        yaml.dump(meta, f, sort_keys=False)\n",
+    "\n",
+    "    print(f\"Finished processing {meta['name']} dataset!\")\n",
+    "\n",
+    "\n",
+    "if __name__ == \"__main__\":\n",
+    "    get_and_transform_data()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "id": "d0474f26-70f3-4655-b81a-df4ada90e7a6",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Found local copy...\n",
+      "Loading...\n",
+      "Done!\n",
+      "Finished processing tcr_epitope_binding dataset!\n"
+     ]
+    }
+   ],
+   "source": [
+    "!python3 transform.py"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "id": "953e7bee-bd5e-41d0-a2be-506e0bc97727",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "total 61592\r\n",
+      "drwxr-xr-x  3 cody  staff    96B Mar  4 10:09 \u001b[34mdata\u001b[m\u001b[m/\r\n",
+      "-rw-r--r--  1 cody  staff    15M Mar  4 10:10 data_clean.csv\r\n",
+      "-rw-r--r--  1 cody  staff    15M Mar  4 10:10 data_original.csv\r\n",
+      "-rw-r--r--  1 cody  staff    45K Mar  4 10:06 example_processing_and_templates.ipynb\r\n",
+      "-rw-r--r--@ 1 cody  staff   1.5K Mar  4 10:10 meta.yaml\r\n",
+      "-rw-r--r--@ 1 cody  staff   4.8K Mar  4 10:10 transform.py\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "ls -lh  # fmt: skip"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "0b08ed06-ba66-4f76-bde1-368ea77d1739",
+   "metadata": {},
+   "source": [
+    "# End"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.9"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/data/tcr_epitope_binding/meta.yaml b/data/tcr_epitope_binding/meta.yaml
new file mode 100644
index 000000000..47b52ccd0
--- /dev/null
+++ b/data/tcr_epitope_binding/meta.yaml
@@ -0,0 +1,60 @@
+name: tcr_epitope_binding
+description: |-
+    T-cells are an integral part of the adaptive immune system, whose survival, proliferation,
+            activationand function are all governed by the interaction of their T-cell receptor (TCR) with immunogenic
+            peptides (epitopes).A large repertoire of T-cell receptors with different specificity is needed to provide
+            protection against a wide range of pathogens.This new task aims to predict the binding affinity
+            given a pair of TCR sequence and epitope sequence.
+targets:
+    - id: binding
+      description: TCR epitope binding.
+      units: ''
+      type: boolean
+      names:
+          - tcr binding affinity
+          - binding affinity
+          - binding
+          - epitope binding affinity
+          - epitope binding
+identifiers:
+    - id: epitope_smiles
+      type: SMILES
+      description: epitope smiles
+    - id: epitope_aa
+      type: Other
+      names:
+          - epitope amino acid sequence
+      description: epitope amino acid sequence
+    - id: tcr
+      type: Other
+      names:
+          - hypervariable CDR3 loop
+      description: hypervariable CDR3 loop
+    - id: tcr_full
+      type: Other
+      names:
+          - tcr full amino acid sequence
+      description: tcr full amino acid sequence
+license: CC BY 4.0
+links:
+    - url: https://tdcommons.ai/multi_pred_tasks/tcrepitope/
+      description: original data set link
+    - url: https://doi.org/10.1093/bioinformatics/btab294
+      description: corresponding publication
+benchmarks:
+    - name: TDC
+      link: https://tdcommons.ai/
+      split_column: split
+num_points: 47182
+bibtex:
+    - |-
+      @article{weber2021titan,
+      title={TITAN: T-cell receptor specificity prediction with bimodal attention network},
+      author={Weber Anna,Born Janis, Martinez Maria Rodriguez},
+      journal={Bioinformatics},
+      volume={56},
+      number={4},
+      pages={i237-i234},
+      year={2021},
+      publisher={Oxford Academic}
+      }
diff --git a/data/tcr_epitope_binding/transform.py b/data/tcr_epitope_binding/transform.py
new file mode 100644
index 000000000..947bf53e2
--- /dev/null
+++ b/data/tcr_epitope_binding/transform.py
@@ -0,0 +1,201 @@
+import pandas as pd
+import yaml
+from tdc.multi_pred import TCREpitopeBinding
+
+
+def get_and_transform_data():
+    # get raw data
+    data = TCREpitopeBinding(name="weber", path="./data")
+
+    split = data.get_split()
+    df_train = split["train"]
+    df_valid = split["valid"]
+    df_test = split["test"]
+    df_train["split"] = "train"
+    df_valid["split"] = "valid"
+    df_test["split"] = "test"
+    df = pd.concat([df_train, df_valid, df_test], axis=0)
+
+    # create dataframenot necessary but ensure we can load the saved data
+
+    # check if fields are the same
+    fields_orig = df.columns.tolist()
+    assert fields_orig == [
+        "epitope_aa",
+        "epitope_smi",
+        "tcr",
+        "tcr_full",
+        "label",
+        "split",
+    ]
+
+    # overwrite column names = fields
+    fields_clean = [
+        "epitope_aa",
+        "epitope_smiles",
+        "tcr",
+        "tcr_full",
+        "binding",
+        "split",
+    ]
+    df.columns = fields_clean
+
+    # data cleaning
+    df.epitope_aa = (
+        df.epitope_aa.str.strip()
+    )  # remove leading and trailing white space characters
+
+    assert not df.duplicated().sum()
+
+    # save to csv
+    fn_data_csv = "data_clean.csv"
+    df.to_csv(fn_data_csv, index=False)
+
+    # create meta yaml
+    meta = {
+        "name": "tcr_epitope_binding",  # unique identifier, we will also use this for directory names
+        "description": """T-cells are an integral part of the adaptive immune system, whose survival, proliferation,
+        activationand function are all governed by the interaction of their T-cell receptor (TCR) with immunogenic
+        peptides (epitopes).A large repertoire of T-cell receptors with different specificity is needed to provide
+        protection against a wide range of pathogens.This new task aims to predict the binding affinity
+        given a pair of TCR sequence and epitope sequence.""",
+        "targets": [
+            {
+                "id": "binding",  # name of the column in a tabular dataset
+                "description": "TCR epitope binding.",  # description of what this column means
+                "units": "",  # units of the values in this column (leave empty if unitless)
+                "type": "boolean",  # can be "categorical", "ordinal", "continuous"
+                "names": [  # names for the property (to sample from for building the prompts)
+                    "tcr binding affinity",
+                    "binding affinity",
+                    "binding",
+                    "epitope binding affinity",
+                    "epitope binding",
+                ],
+            },
+        ],
+        "identifiers": [
+            {
+                "id": "epitope_smiles",  # column name
+                "type": "SMILES",  # can be "SMILES", "SELFIES", "IUPAC", "OTHER"
+                "description": "epitope smiles",  # description (optional, except for "OTHER")
+            },
+            {
+                "id": "epitope_aa",
+                "type": "Other",
+                "names": [
+                    "epitope amino acid sequence",
+                ],
+                "description": "epitope amino acid sequence",
+            },
+            {
+                "id": "tcr",
+                "type": "Other",
+                "names": [
+                    "hypervariable CDR3 loop",
+                ],
+                "description": "hypervariable CDR3 loop",
+            },
+            {
+                "id": "tcr_full",
+                "type": "Other",
+                "names": [
+                    "tcr full amino acid sequence",
+                ],
+                "description": "tcr full amino acid sequence",
+            },
+        ],
+        "templates": [
+            {
+                "prompt": "Please answer the following question.\nPredict the <expt_value_text> for <molecule>.",
+                "completion": "<exp_value>"
+            }
+        ],
+        "fields": [
+            {
+                "exp_value": [
+                 "values": [ 
+                {
+                    "name": "exp_value",
+                    "column": "exp_value",
+                    "text": "binding"
+                }
+                ]
+                ],
+                "molecule": [
+                "values": [
+                {
+                    "name": "epitope_smiles",
+                    "column": "epitope_smiles",
+                    "text": "SMILES"
+
+                },
+                {
+                    "name": "tcr_full",
+                    "column": "tcr_full",
+                    "text":
+
+                },
+
+            ]
+            ],
+
+            }
+        ],
+
+        
+        "license": "CC BY 4.0",  # license under which the original dataset was published
+        "links": [  # list of relevant links (original dataset, other uses, etc.)
+            {
+                "url": "https://tdcommons.ai/multi_pred_tasks/tcrepitope/",
+                "description": "original data set link",
+            },
+            {
+                "url": "https://doi.org/10.1093/bioinformatics/btab294",
+                "description": "corresponding publication",
+            },
+        ],
+        "benchmarks": [
+            {
+                "name": "TDC",
+                "link": "https://tdcommons.ai/",
+                "split_column": "split",
+            },
+        ],
+        "num_points": len(df),  # number of datapoints in this dataset
+        "bibtex": [
+            """@article{weber2021titan,
+title={TITAN: T-cell receptor specificity prediction with bimodal attention network},
+author={Weber Anna,Born Janis, Martinez Maria Rodriguez},
+journal={Bioinformatics},
+volume={56},
+number={4},
+pages={i237-i234},
+year={2021},
+publisher={Oxford Academic}
+}""",
+        ],
+    }
+
+    def str_presenter(dumper, data):
+        """configures yaml for dumping multiline strings
+        Ref: https://stackoverflow.com/questions/8640959/how-can-i-control-what-scalar-form-pyyaml-uses-for-my-data
+        """
+
+        if data.count("\n") > 0:  # check for multiline string
+            return dumper.represent_scalar("tag:yaml.org,2002:str", data, style="|")
+        return dumper.represent_scalar("tag:yaml.org,2002:str", data)
+
+    yaml.add_representer(str, str_presenter)
+    yaml.representer.SafeRepresenter.add_representer(
+        str, str_presenter
+    )  # to use with safe_dum
+    fn_meta = "meta.yaml"
+    with open(fn_meta, "w") as f:
+        yaml.dump(meta, f, sort_keys=False)
+
+    print(f"Finished processing {meta['name']} dataset!")
+
+
+if __name__ == "__main__":
+    get_and_transform_data()