structflo-ner 0.3.0__tar.gz → 0.4.0__tar.gz

structflo_ner-0.4.0/.claude/settings.local.json

@@ -0,0 +1,13 @@
+{
+  "permissions": {
+    "allow": [
+      "Bash(uv sync:*)",
+      "Bash(uv run pytest:*)",
+      "Bash(uv run ruff check:*)",
+      "Bash(uv run ruff format:*)",
+      "Bash(uv run ruff:*)",
+      "Bash(python -m pytest:*)",
+      "Bash(uv run python:*)"
+    ]
+  }
+}

{structflo_ner-0.3.0 → structflo_ner-0.4.0}/.gitignore

@@ -32,3 +32,7 @@ Thumbs.db
 # Environment variables
 .env
 .env.*
+
+
+notebooks/.virtual_documents/
+notebooks/.ipynb_checkpoints/

{structflo_ner-0.3.0 → structflo_ner-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: structflo-ner
-Version: 0.3.0
+Version: 0.4.0
 Summary: Drug discovery NER wrapper around LangExtract — zero-config entity extraction for chemistry and biology.
 License: Apache-2.0
 Requires-Python: >=3.10

{structflo_ner-0.3.0 → structflo_ner-0.4.0}/notebooks/01_quickstart.ipynb

@@ -28,7 +28,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -37,7 +37,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -89,7 +89,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -101,19 +101,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
-       "<div id=\"ner-27203dd1\" style=\"font-family:system-ui,-apple-system,sans-serif;max-width:900px;padding:20px\"><div style=\"display:flex;flex-wrap:wrap;gap:6px;margin-bottom:12px\"><button class=\"ner-legend-btn\" data-target=\"ChemicalEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #3b82f6;border-radius:16px;background:#dbeafe;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#3b82f6\"></span>Compound</button><button class=\"ner-legend-btn\" data-target=\"TargetEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #22c55e;border-radius:16px;background:#dcfce7;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#22c55e\"></span>Target</button><button class=\"ner-legend-btn\" data-target=\"DiseaseEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #ec4899;border-radius:16px;background:#fce7f3;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#ec4899\"></span>Disease</button><button class=\"ner-legend-btn\" data-target=\"BioactivityEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #f59e0b;border-radius:16px;background:#fef3c7;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#f59e0b\"></span>Bioactivity</button></div><div style=\"font-size:0.8em;color:#64748b;margin-bottom:10px\"><b>3</b> Compound · <b>1</b> Target · <b>1</b> Disease · <b>1</b> Bioactivity</div><div style=\"line-height:2;font-size:1em;white-space:pre-wrap\"><mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: compound_name | synonyms: ZD1839\">Gefitinib<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">compound_name</span></mark> (<mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: compound_name\">ZD1839<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">compound_name</span></mark>) is a first-generation <mark class=\"ner-ent\" data-category=\"TargetEntity\" style=\"background:#dcfce7;border-bottom:2px solid #22c55e;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: target\">EGFR<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#22c55e\">target</span></mark> inhibitor with <mark class=\"ner-ent\" data-category=\"BioactivityEntity\" style=\"background:#fef3c7;border-bottom:2px solid #f59e0b;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: bioactivity | value: 0.033 | unit: µM | assay_type: IC50\">IC50 = 0.033 µM<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#f59e0b\">bioactivity</span></mark> approved for <mark class=\"ner-ent\" data-category=\"DiseaseEntity\" style=\"background:#fce7f3;border-bottom:2px solid #ec4899;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: disease | therapeutic_area: oncology\">NSCLC<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#ec4899\">disease</span></mark>.Its SMILES is <mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: smiles\">COc1cc2ncnc(Nc3ccc(F)c(Cl)c3)c2cc1OCCCN1CCOCC1<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">smiles</span></mark>.</div><script>(function(){var w=document.getElementById('ner-27203dd1');w.querySelectorAll('.ner-legend-btn').forEach(function(btn){btn.addEventListener('click',function(){var cat=btn.dataset.target;var active=btn.dataset.active!=='false';btn.dataset.active=active?'false':'true';btn.style.opacity=active?'0.35':'1';w.querySelectorAll('[data-category=\"'+cat+'\"]').forEach(function(el){if(el.classList.contains('ner-ent'))el.style.background=active?'transparent':el.style.borderBottomColor.replace('solid ','');el.style.opacity=active?'0.3':'1';});});});})()</script></div>"
+       "<div id=\"ner-e5927285\" style=\"font-family:system-ui,-apple-system,sans-serif;max-width:900px;padding:20px\"><div style=\"display:flex;flex-wrap:wrap;gap:6px;margin-bottom:12px\"><button class=\"ner-legend-btn\" data-target=\"ChemicalEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #3b82f6;border-radius:16px;background:#dbeafe;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#3b82f6\"></span>Compound</button><button class=\"ner-legend-btn\" data-target=\"TargetEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #22c55e;border-radius:16px;background:#dcfce7;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#22c55e\"></span>Target</button><button class=\"ner-legend-btn\" data-target=\"DiseaseEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #ec4899;border-radius:16px;background:#fce7f3;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#ec4899\"></span>Disease</button><button class=\"ner-legend-btn\" data-target=\"BioactivityEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #f59e0b;border-radius:16px;background:#fef3c7;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#f59e0b\"></span>Bioactivity</button></div><div style=\"font-size:0.8em;color:#64748b;margin-bottom:10px\"><b>3</b> Compound · <b>1</b> Target · <b>1</b> Disease · <b>1</b> Bioactivity</div><div style=\"line-height:2;font-size:1em;white-space:pre-wrap\"><mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: compound_name | synonyms: ZD1839\">Gefitinib<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">compound_name</span></mark> (<mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: compound_name\">ZD1839<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">compound_name</span></mark>) is a first-generation <mark class=\"ner-ent\" data-category=\"TargetEntity\" style=\"background:#dcfce7;border-bottom:2px solid #22c55e;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: target\">EGFR<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#22c55e\">target</span></mark> inhibitor with <mark class=\"ner-ent\" data-category=\"BioactivityEntity\" style=\"background:#fef3c7;border-bottom:2px solid #f59e0b;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: bioactivity | value: 0.033 | unit: µM | assay_type: IC50 | compound_name: Gefitinib\">IC50 = 0.033 µM<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#f59e0b\">bioactivity</span></mark> approved for <mark class=\"ner-ent\" data-category=\"DiseaseEntity\" style=\"background:#fce7f3;border-bottom:2px solid #ec4899;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: disease | therapeutic_area: oncology\">NSCLC<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#ec4899\">disease</span></mark>.Its SMILES is <mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: smiles\">COc1cc2ncnc(Nc3ccc(F)c(Cl)c3)c2cc1OCCCN1CCOCC1<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">smiles</span></mark>.</div><script>(function(){var w=document.getElementById('ner-e5927285');w.querySelectorAll('.ner-legend-btn').forEach(function(btn){btn.addEventListener('click',function(){var cat=btn.dataset.target;var active=btn.dataset.active!=='false';btn.dataset.active=active?'false':'true';btn.style.opacity=active?'0.35':'1';w.querySelectorAll('[data-category=\"'+cat+'\"]').forEach(function(el){if(el.classList.contains('ner-ent'))el.style.background=active?'transparent':el.style.borderBottomColor.replace('solid ','');el.style.opacity=active?'0.3':'1';});});});})()</script></div>"
       ],
       "text/plain": [
-       "NERResult(source_text='Gefitinib (ZD1839) is a first-generation EGFR inhibitor with IC50 = 0.033 µM approved for NSCLC.Its SMILES is COc1cc2ncnc(Nc3ccc(F)c(Cl)c3)c2cc1OCCCN1CCOCC1.', compounds=[ChemicalEntity(text='Gefitinib', entity_type='compound_name', char_start=0, char_end=9, attributes={'synonyms': 'ZD1839'}, alignment='match_exact'), ChemicalEntity(text='ZD1839', entity_type='compound_name', char_start=11, char_end=17, attributes={}, alignment='match_exact'), ChemicalEntity(text='COc1cc2ncnc(Nc3ccc(F)c(Cl)c3)c2cc1OCCCN1CCOCC1', entity_type='smiles', char_start=110, char_end=156, attributes={}, alignment='match_exact')], targets=[TargetEntity(text='EGFR', entity_type='target', char_start=41, char_end=45, attributes={}, alignment='match_exact')], diseases=[DiseaseEntity(text='NSCLC', entity_type='disease', char_start=90, char_end=95, attributes={'therapeutic_area': 'oncology'}, alignment='match_exact')], bioactivities=[BioactivityEntity(text='IC50 = 0.033 µM', entity_type='bioactivity', char_start=61, char_end=76, attributes={'value': '0.033', 'unit': 'µM', 'assay_type': 'IC50'}, alignment='match_exact')], assays=[], mechanisms=[], accessions=[], products=[], functional_categories=[], screening_methods=[], unclassified=[])"
+       "NERResult(source_text='Gefitinib (ZD1839) is a first-generation EGFR inhibitor with IC50 = 0.033 µM approved for NSCLC.Its SMILES is COc1cc2ncnc(Nc3ccc(F)c(Cl)c3)c2cc1OCCCN1CCOCC1.', compounds=[ChemicalEntity(text='Gefitinib', entity_type='compound_name', char_start=0, char_end=9, attributes={'synonyms': 'ZD1839'}, alignment='match_exact'), ChemicalEntity(text='ZD1839', entity_type='compound_name', char_start=11, char_end=17, attributes={}, alignment='match_exact'), ChemicalEntity(text='COc1cc2ncnc(Nc3ccc(F)c(Cl)c3)c2cc1OCCCN1CCOCC1', entity_type='smiles', char_start=110, char_end=156, attributes={}, alignment='match_exact')], targets=[TargetEntity(text='EGFR', entity_type='target', char_start=41, char_end=45, attributes={}, alignment='match_exact')], diseases=[DiseaseEntity(text='NSCLC', entity_type='disease', char_start=90, char_end=95, attributes={'therapeutic_area': 'oncology'}, alignment='match_exact')], bioactivities=[BioactivityEntity(text='IC50 = 0.033 µM', entity_type='bioactivity', char_start=61, char_end=76, attributes={'value': '0.033', 'unit': 'µM', 'assay_type': 'IC50', 'compound_name': 'Gefitinib'}, alignment='match_exact')], assays=[], mechanisms=[], accessions=[], products=[], functional_categories=[], screening_methods=[], strains=[], unclassified=[])"
       ]
      },
-     "execution_count": 32,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -127,7 +127,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -162,6 +162,7 @@
        "      <th>value</th>\n",
        "      <th>unit</th>\n",
        "      <th>assay_type</th>\n",
+       "      <th>compound_name</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
@@ -178,6 +179,7 @@
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
@@ -192,6 +194,7 @@
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
@@ -206,6 +209,7 @@
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
@@ -220,6 +224,7 @@
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
@@ -234,6 +239,7 @@
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>5</th>\n",
@@ -248,6 +254,7 @@
        "      <td>0.033</td>\n",
        "      <td>µM</td>\n",
        "      <td>IC50</td>\n",
+       "      <td>Gefitinib</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
@@ -270,16 +277,16 @@
        "4      DiseaseEntity          90        95  match_exact      NaN   \n",
        "5  BioactivityEntity          61        76  match_exact      NaN   \n",
        "\n",
-       "  therapeutic_area  value unit assay_type  \n",
-       "0              NaN    NaN  NaN        NaN  \n",
-       "1              NaN    NaN  NaN        NaN  \n",
-       "2              NaN    NaN  NaN        NaN  \n",
-       "3              NaN    NaN  NaN        NaN  \n",
-       "4         oncology    NaN  NaN        NaN  \n",
-       "5              NaN  0.033   µM       IC50  "
+       "  therapeutic_area  value unit assay_type compound_name  \n",
+       "0              NaN    NaN  NaN        NaN           NaN  \n",
+       "1              NaN    NaN  NaN        NaN           NaN  \n",
+       "2              NaN    NaN  NaN        NaN           NaN  \n",
+       "3              NaN    NaN  NaN        NaN           NaN  \n",
+       "4         oncology    NaN  NaN        NaN           NaN  \n",
+       "5              NaN  0.033   µM       IC50     Gefitinib  "
       ]
      },
-     "execution_count": 33,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -311,19 +318,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/html": [
-       "<div id=\"ner-5d2cb6c8\" style=\"font-family:system-ui,-apple-system,sans-serif;max-width:900px;padding:20px\"><div style=\"display:flex;flex-wrap:wrap;gap:6px;margin-bottom:12px\"><button class=\"ner-legend-btn\" data-target=\"ChemicalEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #3b82f6;border-radius:16px;background:#dbeafe;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#3b82f6\"></span>Compound</button><button class=\"ner-legend-btn\" data-target=\"TargetEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #22c55e;border-radius:16px;background:#dcfce7;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#22c55e\"></span>Target</button><button class=\"ner-legend-btn\" data-target=\"DiseaseEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #ec4899;border-radius:16px;background:#fce7f3;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#ec4899\"></span>Disease</button><button class=\"ner-legend-btn\" data-target=\"AccessionEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #14b8a6;border-radius:16px;background:#ccfbf1;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#14b8a6\"></span>Accession</button><button class=\"ner-legend-btn\" data-target=\"FunctionalCategoryEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #f43f5e;border-radius:16px;background:#ffe4e6;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#f43f5e\"></span>Function</button><button class=\"ner-legend-btn\" data-target=\"ScreeningMethodEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #10b981;border-radius:16px;background:#d1fae5;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#10b981\"></span>Screening</button></div><div style=\"font-size:0.8em;color:#64748b;margin-bottom:10px\"><b>2</b> Compound · <b>1</b> Target · <b>2</b> Disease · <b>1</b> Accession · <b>1</b> Function · <b>1</b> Screening</div><div style=\"line-height:2;font-size:1em;white-space:pre-wrap\"><mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: compound_name | synonyms: TMC207\">Bedaquiline<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">compound_name</span></mark> (<mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: compound_name\">TMC207<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">compound_name</span></mark>) is a diarylquinoline that inhibits the mycobacterial <mark class=\"ner-ent\" data-category=\"TargetEntity\" style=\"background:#dcfce7;border-bottom:2px solid #22c55e;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: target | gene_name: atpE | protein_family: ATP synthase\">ATP synthase subunit c<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#22c55e\">target</span></mark> encoded by atpE (<mark class=\"ner-ent\" data-category=\"AccessionEntity\" style=\"background:#ccfbf1;border-bottom:2px solid #14b8a6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: accession_number\">Rv1305<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#14b8a6\">accession_number</span></mark>). It shows potent activity against Mycobacterium tuberculosis including <mark class=\"ner-ent\" data-category=\"DiseaseEntity\" style=\"background:#fce7f3;border-bottom:2px solid #ec4899;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: disease | therapeutic_area: infectious disease\">MDR-TB<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#ec4899\">disease</span></mark> and <mark class=\"ner-ent\" data-category=\"DiseaseEntity\" style=\"background:#fce7f3;border-bottom:2px solid #ec4899;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: disease | therapeutic_area: infectious disease\">XDR-TB<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#ec4899\">disease</span></mark>. This compound was identified through <mark class=\"ner-ent\" data-category=\"ScreeningMethodEntity\" style=\"background:#d1fae5;border-bottom:2px solid #10b981;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: screening_method\">whole-cell screening<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#10b981\">screening_method</span></mark> and targets the <mark class=\"ner-ent\" data-category=\"FunctionalCategoryEntity\" style=\"background:#ffe4e6;border-bottom:2px solid #f43f5e;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: functional_category\">energy metabolism pathway<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#f43f5e\">functional_category</span></mark>.</div><script>(function(){var w=document.getElementById('ner-5d2cb6c8');w.querySelectorAll('.ner-legend-btn').forEach(function(btn){btn.addEventListener('click',function(){var cat=btn.dataset.target;var active=btn.dataset.active!=='false';btn.dataset.active=active?'false':'true';btn.style.opacity=active?'0.35':'1';w.querySelectorAll('[data-category=\"'+cat+'\"]').forEach(function(el){if(el.classList.contains('ner-ent'))el.style.background=active?'transparent':el.style.borderBottomColor.replace('solid ','');el.style.opacity=active?'0.3':'1';});});});})()</script></div>"
+       "<div id=\"ner-742315f5\" style=\"font-family:system-ui,-apple-system,sans-serif;max-width:900px;padding:20px\"><div style=\"display:flex;flex-wrap:wrap;gap:6px;margin-bottom:12px\"><button class=\"ner-legend-btn\" data-target=\"ChemicalEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #3b82f6;border-radius:16px;background:#dbeafe;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#3b82f6\"></span>Compound</button><button class=\"ner-legend-btn\" data-target=\"TargetEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #22c55e;border-radius:16px;background:#dcfce7;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#22c55e\"></span>Target</button><button class=\"ner-legend-btn\" data-target=\"DiseaseEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #ec4899;border-radius:16px;background:#fce7f3;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#ec4899\"></span>Disease</button><button class=\"ner-legend-btn\" data-target=\"BioactivityEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #f59e0b;border-radius:16px;background:#fef3c7;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#f59e0b\"></span>Bioactivity</button><button class=\"ner-legend-btn\" data-target=\"AccessionEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #14b8a6;border-radius:16px;background:#ccfbf1;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#14b8a6\"></span>Accession</button><button class=\"ner-legend-btn\" data-target=\"FunctionalCategoryEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #f43f5e;border-radius:16px;background:#ffe4e6;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#f43f5e\"></span>Function</button><button class=\"ner-legend-btn\" data-target=\"ScreeningMethodEntity\" style=\"display:inline-flex;align-items:center;gap:4px;padding:4px 10px;border:1.5px solid #10b981;border-radius:16px;background:#d1fae5;cursor:pointer;font-size:0.8em;font-weight:500;font-family:inherit\"><span style=\"width:8px;height:8px;border-radius:50%;background:#10b981\"></span>Screening</button></div><div style=\"font-size:0.8em;color:#64748b;margin-bottom:10px\"><b>2</b> Compound · <b>1</b> Target · <b>2</b> Disease · <b>1</b> Bioactivity · <b>1</b> Accession · <b>1</b> Function · <b>1</b> Screening</div><div style=\"line-height:2;font-size:1em;white-space:pre-wrap\"><mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: compound_name | synonyms: TMC207\">Bedaquiline<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">compound_name</span></mark> (<mark class=\"ner-ent\" data-category=\"ChemicalEntity\" style=\"background:#dbeafe;border-bottom:2px solid #3b82f6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: compound_name\">TMC207<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#3b82f6\">compound_name</span></mark>) is a diarylquinoline that inhibits the mycobacterial <mark class=\"ner-ent\" data-category=\"TargetEntity\" style=\"background:#dcfce7;border-bottom:2px solid #22c55e;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: target | gene_name: atpE | protein_family: ATP synthase\">ATP synthase subunit c<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#22c55e\">target</span></mark> encoded by atpE (<mark class=\"ner-ent\" data-category=\"AccessionEntity\" style=\"background:#ccfbf1;border-bottom:2px solid #14b8a6;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: accession_number\">Rv1305<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#14b8a6\">accession_number</span></mark>) with <mark class=\"ner-ent\" data-category=\"BioactivityEntity\" style=\"background:#fef3c7;border-bottom:2px solid #f59e0b;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: bioactivity | value: 0.033 | unit: µM | assay_type: IC50 | compound_name: Bedaquiline\">IC50 = 0.033<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#f59e0b\">bioactivity</span></mark> µMIt shows potent activity against Mycobacterium tuberculosis including <mark class=\"ner-ent\" data-category=\"DiseaseEntity\" style=\"background:#fce7f3;border-bottom:2px solid #ec4899;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: disease | therapeutic_area: infectious disease\">MDR-TB<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#ec4899\">disease</span></mark> and <mark class=\"ner-ent\" data-category=\"DiseaseEntity\" style=\"background:#fce7f3;border-bottom:2px solid #ec4899;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: disease | therapeutic_area: infectious disease\">XDR-TB<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#ec4899\">disease</span></mark>. This compound was identified through <mark class=\"ner-ent\" data-category=\"ScreeningMethodEntity\" style=\"background:#d1fae5;border-bottom:2px solid #10b981;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: screening_method\">whole-cell screening<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#10b981\">screening_method</span></mark> and targets the <mark class=\"ner-ent\" data-category=\"FunctionalCategoryEntity\" style=\"background:#ffe4e6;border-bottom:2px solid #f43f5e;padding:2px 4px;border-radius:4px;cursor:default\" title=\"type: functional_category\">energy metabolism pathway<span style=\"font-size:0.7em;font-weight:600;vertical-align:super;margin-left:2px;color:#f43f5e\">functional_category</span></mark>.</div><script>(function(){var w=document.getElementById('ner-742315f5');w.querySelectorAll('.ner-legend-btn').forEach(function(btn){btn.addEventListener('click',function(){var cat=btn.dataset.target;var active=btn.dataset.active!=='false';btn.dataset.active=active?'false':'true';btn.style.opacity=active?'0.35':'1';w.querySelectorAll('[data-category=\"'+cat+'\"]').forEach(function(el){if(el.classList.contains('ner-ent'))el.style.background=active?'transparent':el.style.borderBottomColor.replace('solid ','');el.style.opacity=active?'0.3':'1';});});});})()</script></div>"
       ],
       "text/plain": [
-       "NERResult(source_text='Bedaquiline (TMC207) is a diarylquinoline that inhibits the mycobacterial ATP synthase subunit c encoded by atpE (Rv1305). It shows potent activity against Mycobacterium tuberculosis including MDR-TB and XDR-TB. This compound was identified through whole-cell screening and targets the energy metabolism pathway.', compounds=[ChemicalEntity(text='Bedaquiline', entity_type='compound_name', char_start=0, char_end=11, attributes={'synonyms': 'TMC207'}, alignment='match_exact'), ChemicalEntity(text='TMC207', entity_type='compound_name', char_start=13, char_end=19, attributes={}, alignment='match_exact')], targets=[TargetEntity(text='ATP synthase subunit c', entity_type='target', char_start=74, char_end=96, attributes={'gene_name': 'atpE', 'protein_family': 'ATP synthase'}, alignment='match_exact')], diseases=[DiseaseEntity(text='MDR-TB', entity_type='disease', char_start=193, char_end=199, attributes={'therapeutic_area': 'infectious disease'}, alignment='match_exact'), DiseaseEntity(text='XDR-TB', entity_type='disease', char_start=204, char_end=210, attributes={'therapeutic_area': 'infectious disease'}, alignment='match_exact')], bioactivities=[], assays=[], mechanisms=[], accessions=[AccessionEntity(text='Rv1305', entity_type='accession_number', char_start=114, char_end=120, attributes={}, alignment='match_exact')], products=[], functional_categories=[FunctionalCategoryEntity(text='energy metabolism pathway', entity_type='functional_category', char_start=286, char_end=311, attributes={}, alignment='match_exact')], screening_methods=[ScreeningMethodEntity(text='whole-cell screening', entity_type='screening_method', char_start=249, char_end=269, attributes={}, alignment='match_exact')], unclassified=[])"
+       "NERResult(source_text='Bedaquiline (TMC207) is a diarylquinoline that inhibits the mycobacterial ATP synthase subunit c encoded by atpE (Rv1305) with IC50 = 0.033 µMIt shows potent activity against Mycobacterium tuberculosis including MDR-TB and XDR-TB. This compound was identified through whole-cell screening and targets the energy metabolism pathway.', compounds=[ChemicalEntity(text='Bedaquiline', entity_type='compound_name', char_start=0, char_end=11, attributes={'synonyms': 'TMC207'}, alignment='match_exact'), ChemicalEntity(text='TMC207', entity_type='compound_name', char_start=13, char_end=19, attributes={}, alignment='match_exact')], targets=[TargetEntity(text='ATP synthase subunit c', entity_type='target', char_start=74, char_end=96, attributes={'gene_name': 'atpE', 'protein_family': 'ATP synthase'}, alignment='match_exact')], diseases=[DiseaseEntity(text='MDR-TB', entity_type='disease', char_start=212, char_end=218, attributes={'therapeutic_area': 'infectious disease'}, alignment='match_exact'), DiseaseEntity(text='XDR-TB', entity_type='disease', char_start=223, char_end=229, attributes={'therapeutic_area': 'infectious disease'}, alignment='match_exact')], bioactivities=[BioactivityEntity(text='IC50 = 0.033 µM', entity_type='bioactivity', char_start=127, char_end=139, attributes={'value': '0.033', 'unit': 'µM', 'assay_type': 'IC50', 'compound_name': 'Bedaquiline'}, alignment='match_lesser')], assays=[], mechanisms=[], accessions=[AccessionEntity(text='Rv1305', entity_type='accession_number', char_start=114, char_end=120, attributes={}, alignment='match_exact')], products=[], functional_categories=[FunctionalCategoryEntity(text='energy metabolism pathway', entity_type='functional_category', char_start=305, char_end=330, attributes={}, alignment='match_exact')], screening_methods=[ScreeningMethodEntity(text='whole-cell screening', entity_type='screening_method', char_start=268, char_end=288, attributes={}, alignment='match_exact')], strains=[], unclassified=[])"
       ]
      },
-     "execution_count": 10,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -331,7 +338,7 @@
    "source": [
     "text = (\n",
     "    \"Bedaquiline (TMC207) is a diarylquinoline that inhibits the \"\n",
-    "    \"mycobacterial ATP synthase subunit c encoded by atpE (Rv1305). \"\n",
+    "    \"mycobacterial ATP synthase subunit c encoded by atpE (Rv1305) with IC50 = 0.033 µM\"\n",
     "    \"It shows potent activity against Mycobacterium tuberculosis \"\n",
     "    \"including MDR-TB and XDR-TB. This compound was identified through \"\n",
     "    \"whole-cell screening and targets the energy metabolism pathway.\"\n",
@@ -340,126 +347,9 @@
     "result"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## 3. Built-in profiles\n",
-    "\n",
-    "Profiles control which entity types are extracted. Use them to focus the model on specific categories."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Compounds: [ChemicalEntity(text='Bedaquiline', entity_type='compound_name', char_start=0, char_end=11, attributes={'synonyms': 'TMC207'}, alignment='match_exact'), ChemicalEntity(text='TMC207', entity_type='compound_name', char_start=13, char_end=19, attributes={}, alignment='match_exact')]\n",
-      "Targets: []\n"
-     ]
-    }
-   ],
-   "source": [
-    "from structflo.ner import CHEMISTRY, BIOLOGY\n",
-    "# Extract only chemical entities\n",
-    "chem_result = extractor.extract(text, profile=CHEMISTRY)\n",
-    "print(\"Compounds:\", chem_result.compounds)\n",
-    "print(\"Targets:\", chem_result.targets)  # empty — not in CHEMISTRY profile"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Profile: chemistry+biology\n",
-      "Entity classes: ['compound_name', 'smiles', 'cas_number', 'molecular_formula', 'target', 'gene_name', 'protein_name']\n",
-      "Compounds: [ChemicalEntity(text='Bedaquiline', entity_type='compound_name', char_start=0, char_end=11, attributes={'synonyms': 'TMC207'}, alignment='match_exact'), ChemicalEntity(text='TMC207', entity_type='compound_name', char_start=13, char_end=19, attributes={}, alignment='match_exact')]\n",
-      "Targets: [TargetEntity(text='mycobacterial ATP synthase subunit c', entity_type='target', char_start=60, char_end=96, attributes={'gene_name': 'atpE (Rv1305)', 'organism': 'Mycobacterium tuberculosis'}, alignment='match_exact'), TargetEntity(text='energy metabolism pathway', entity_type='target', char_start=286, char_end=311, attributes={'protein_family': 'pathway', 'organism': 'Mycobacterium tuberculosis'}, alignment='match_exact')]\n"
-     ]
-    }
-   ],
-   "source": [
-    "# Merge profiles to combine entity types\n",
-    "combined = CHEMISTRY.merge(BIOLOGY)\n",
-    "print(f\"Profile: {combined.name}\")\n",
-    "print(f\"Entity classes: {combined.entity_classes}\")\n",
-    "\n",
-    "combined_result = extractor.extract(text, profile=combined)\n",
-    "print(\"Compounds:\", combined_result.compounds)\n",
-    "print(\"Targets:\", combined_result.targets)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## 5. Working with results"
-   ]
-  },
   {
    "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Compounds: [ChemicalEntity(text='Bedaquiline', entity_type='compound_name', char_start=0, char_end=11, attributes={'synonyms': 'TMC207'}, alignment='match_exact'), ChemicalEntity(text='TMC207', entity_type='compound_name', char_start=13, char_end=19, attributes={}, alignment='match_exact')]\n",
-      "Targets: [TargetEntity(text='ATP synthase subunit c', entity_type='target', char_start=74, char_end=96, attributes={'gene_name': 'atpE', 'protein_family': 'ATP synthase'}, alignment='match_exact')]\n",
-      "Bioactivities: []\n",
-      "Diseases: [DiseaseEntity(text='MDR-TB', entity_type='disease', char_start=193, char_end=199, attributes={'therapeutic_area': 'infectious disease'}, alignment='match_exact'), DiseaseEntity(text='XDR-TB', entity_type='disease', char_start=204, char_end=210, attributes={'therapeutic_area': 'infectious disease'}, alignment='match_exact')]\n",
-      "Mechanisms: []\n"
-     ]
-    }
-   ],
-   "source": [
-    "# Access typed entity lists\n",
-    "print(\"Compounds:\", result.compounds)\n",
-    "print(\"Targets:\", result.targets)\n",
-    "print(\"Bioactivities:\", result.bioactivities)\n",
-    "print(\"Diseases:\", result.diseases)\n",
-    "print(\"Mechanisms:\", result.mechanisms)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 16,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "compound_name        | Bedaquiline\n",
-      "compound_name        | TMC207\n",
-      "target               | ATP synthase subunit c\n",
-      "disease              | MDR-TB\n",
-      "disease              | XDR-TB\n",
-      "accession_number     | Rv1305\n",
-      "functional_category  | energy metabolism pathway\n",
-      "screening_method     | whole-cell screening\n"
-     ]
-    }
-   ],
-   "source": [
-    "# Flat list of all entities\n",
-    "for entity in result.all_entities():\n",
-    "    print(f\"{entity.entity_type:20s} | {entity.text}\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -493,6 +383,10 @@
        "      <th>gene_name</th>\n",
        "      <th>protein_family</th>\n",
        "      <th>therapeutic_area</th>\n",
+       "      <th>value</th>\n",
+       "      <th>unit</th>\n",
+       "      <th>assay_type</th>\n",
+       "      <th>compound_name</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
@@ -508,6 +402,10 @@
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
@@ -521,6 +419,10 @@
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
@@ -534,35 +436,64 @@
        "      <td>atpE</td>\n",
        "      <td>ATP synthase</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td>MDR-TB</td>\n",
        "      <td>disease</td>\n",
        "      <td>DiseaseEntity</td>\n",
-       "      <td>193</td>\n",
-       "      <td>199</td>\n",
+       "      <td>212</td>\n",
+       "      <td>218</td>\n",
        "      <td>match_exact</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>infectious disease</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td>XDR-TB</td>\n",
        "      <td>disease</td>\n",
        "      <td>DiseaseEntity</td>\n",
-       "      <td>204</td>\n",
-       "      <td>210</td>\n",
+       "      <td>223</td>\n",
+       "      <td>229</td>\n",
        "      <td>match_exact</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>infectious disease</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>5</th>\n",
+       "      <td>IC50 = 0.033 µM</td>\n",
+       "      <td>bioactivity</td>\n",
+       "      <td>BioactivityEntity</td>\n",
+       "      <td>127</td>\n",
+       "      <td>139</td>\n",
+       "      <td>match_lesser</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>0.033</td>\n",
+       "      <td>µM</td>\n",
+       "      <td>IC50</td>\n",
+       "      <td>Bedaquiline</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
        "      <td>Rv1305</td>\n",
        "      <td>accession_number</td>\n",
        "      <td>AccessionEntity</td>\n",
@@ -573,32 +504,44 @@
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>6</th>\n",
+       "      <th>7</th>\n",
        "      <td>energy metabolism pathway</td>\n",
        "      <td>functional_category</td>\n",
        "      <td>FunctionalCategoryEntity</td>\n",
-       "      <td>286</td>\n",
-       "      <td>311</td>\n",
+       "      <td>305</td>\n",
+       "      <td>330</td>\n",
        "      <td>match_exact</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>7</th>\n",
+       "      <th>8</th>\n",
        "      <td>whole-cell screening</td>\n",
        "      <td>screening_method</td>\n",
        "      <td>ScreeningMethodEntity</td>\n",
-       "      <td>249</td>\n",
-       "      <td>269</td>\n",
+       "      <td>268</td>\n",
+       "      <td>288</td>\n",
        "      <td>match_exact</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
@@ -611,36 +554,118 @@
        "2     ATP synthase subunit c               target              TargetEntity   \n",
        "3                     MDR-TB              disease             DiseaseEntity   \n",
        "4                     XDR-TB              disease             DiseaseEntity   \n",
-       "5                     Rv1305     accession_number           AccessionEntity   \n",
-       "6  energy metabolism pathway  functional_category  FunctionalCategoryEntity   \n",
-       "7       whole-cell screening     screening_method     ScreeningMethodEntity   \n",
+       "5            IC50 = 0.033 µM          bioactivity         BioactivityEntity   \n",
+       "6                     Rv1305     accession_number           AccessionEntity   \n",
+       "7  energy metabolism pathway  functional_category  FunctionalCategoryEntity   \n",
+       "8       whole-cell screening     screening_method     ScreeningMethodEntity   \n",
        "\n",
-       "   char_start  char_end    alignment synonyms gene_name protein_family  \\\n",
-       "0           0        11  match_exact   TMC207       NaN            NaN   \n",
-       "1          13        19  match_exact      NaN       NaN            NaN   \n",
-       "2          74        96  match_exact      NaN      atpE   ATP synthase   \n",
-       "3         193       199  match_exact      NaN       NaN            NaN   \n",
-       "4         204       210  match_exact      NaN       NaN            NaN   \n",
-       "5         114       120  match_exact      NaN       NaN            NaN   \n",
-       "6         286       311  match_exact      NaN       NaN            NaN   \n",
-       "7         249       269  match_exact      NaN       NaN            NaN   \n",
+       "   char_start  char_end     alignment synonyms gene_name protein_family  \\\n",
+       "0           0        11   match_exact   TMC207       NaN            NaN   \n",
+       "1          13        19   match_exact      NaN       NaN            NaN   \n",
+       "2          74        96   match_exact      NaN      atpE   ATP synthase   \n",
+       "3         212       218   match_exact      NaN       NaN            NaN   \n",
+       "4         223       229   match_exact      NaN       NaN            NaN   \n",
+       "5         127       139  match_lesser      NaN       NaN            NaN   \n",
+       "6         114       120   match_exact      NaN       NaN            NaN   \n",
+       "7         305       330   match_exact      NaN       NaN            NaN   \n",
+       "8         268       288   match_exact      NaN       NaN            NaN   \n",
        "\n",
-       "     therapeutic_area  \n",
-       "0                 NaN  \n",
-       "1                 NaN  \n",
-       "2                 NaN  \n",
-       "3  infectious disease  \n",
-       "4  infectious disease  \n",
-       "5                 NaN  \n",
-       "6                 NaN  \n",
-       "7                 NaN  "
+       "     therapeutic_area  value unit assay_type compound_name  \n",
+       "0                 NaN    NaN  NaN        NaN           NaN  \n",
+       "1                 NaN    NaN  NaN        NaN           NaN  \n",
+       "2                 NaN    NaN  NaN        NaN           NaN  \n",
+       "3  infectious disease    NaN  NaN        NaN           NaN  \n",
+       "4  infectious disease    NaN  NaN        NaN           NaN  \n",
+       "5                 NaN  0.033   µM       IC50   Bedaquiline  \n",
+       "6                 NaN    NaN  NaN        NaN           NaN  \n",
+       "7                 NaN    NaN  NaN        NaN           NaN  \n",
+       "8                 NaN    NaN  NaN        NaN           NaN  "
       ]
      },
-     "execution_count": 17,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
+   "source": [
+    "result.to_dataframe()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 3. Built-in profiles\n",
+    "\n",
+    "Profiles control which entity types are extracted. Use them to focus the model on specific categories."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from structflo.ner import CHEMISTRY, BIOLOGY\n",
+    "# Extract only chemical entities\n",
+    "chem_result = extractor.extract(text, profile=CHEMISTRY)\n",
+    "print(\"Compounds:\", chem_result.compounds)\n",
+    "print(\"Targets:\", chem_result.targets)  # empty — not in CHEMISTRY profile"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Merge profiles to combine entity types\n",
+    "combined = CHEMISTRY.merge(BIOLOGY)\n",
+    "print(f\"Profile: {combined.name}\")\n",
+    "print(f\"Entity classes: {combined.entity_classes}\")\n",
+    "\n",
+    "combined_result = extractor.extract(text, profile=combined)\n",
+    "print(\"Compounds:\", combined_result.compounds)\n",
+    "print(\"Targets:\", combined_result.targets)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 5. Working with results"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Access typed entity lists\n",
+    "print(\"Compounds:\", result.compounds)\n",
+    "print(\"Targets:\", result.targets)\n",
+    "print(\"Bioactivities:\", result.bioactivities)\n",
+    "print(\"Diseases:\", result.diseases)\n",
+    "print(\"Mechanisms:\", result.mechanisms)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Flat list of all entities\n",
+    "for entity in result.all_entities():\n",
+    "    print(f\"{entity.entity_type:20s} | {entity.text}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "# Export to pandas DataFrame\n",
     "df = result.to_dataframe()\n",
@@ -649,109 +674,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "{\n",
-      "  \"source_text\": \"Bedaquiline (TMC207) is a diarylquinoline that inhibits the mycobacterial ATP synthase subunit c encoded by atpE (Rv1305). It shows potent activity against Mycobacterium tuberculosis including MDR-TB and XDR-TB. This compound was identified through whole-cell screening and targets the energy metabolism pathway.\",\n",
-      "  \"compounds\": [\n",
-      "    {\n",
-      "      \"text\": \"Bedaquiline\",\n",
-      "      \"entity_type\": \"compound_name\",\n",
-      "      \"char_start\": 0,\n",
-      "      \"char_end\": 11,\n",
-      "      \"attributes\": {\n",
-      "        \"synonyms\": \"TMC207\"\n",
-      "      },\n",
-      "      \"alignment\": \"match_exact\"\n",
-      "    },\n",
-      "    {\n",
-      "      \"text\": \"TMC207\",\n",
-      "      \"entity_type\": \"compound_name\",\n",
-      "      \"char_start\": 13,\n",
-      "      \"char_end\": 19,\n",
-      "      \"attributes\": {},\n",
-      "      \"alignment\": \"match_exact\"\n",
-      "    }\n",
-      "  ],\n",
-      "  \"targets\": [\n",
-      "    {\n",
-      "      \"text\": \"ATP synthase subunit c\",\n",
-      "      \"entity_type\": \"target\",\n",
-      "      \"char_start\": 74,\n",
-      "      \"char_end\": 96,\n",
-      "      \"attributes\": {\n",
-      "        \"gene_name\": \"atpE\",\n",
-      "        \"protein_family\": \"ATP synthase\"\n",
-      "      },\n",
-      "      \"alignment\": \"match_exact\"\n",
-      "    }\n",
-      "  ],\n",
-      "  \"diseases\": [\n",
-      "    {\n",
-      "      \"text\": \"MDR-TB\",\n",
-      "      \"entity_type\": \"disease\",\n",
-      "      \"char_start\": 193,\n",
-      "      \"char_end\": 199,\n",
-      "      \"attributes\": {\n",
-      "        \"therapeutic_area\": \"infectious disease\"\n",
-      "      },\n",
-      "      \"alignment\": \"match_exact\"\n",
-      "    },\n",
-      "    {\n",
-      "      \"text\": \"XDR-TB\",\n",
-      "      \"entity_type\": \"disease\",\n",
-      "      \"char_start\": 204,\n",
-      "      \"char_end\": 210,\n",
-      "      \"attributes\": {\n",
-      "        \"therapeutic_area\": \"infectious disease\"\n",
-      "      },\n",
-      "      \"alignment\": \"match_exact\"\n",
-      "    }\n",
-      "  ],\n",
-      "  \"bioactivities\": [],\n",
-      "  \"assays\": [],\n",
-      "  \"mechanisms\": [],\n",
-      "  \"accessions\": [\n",
-      "    {\n",
-      "      \"text\": \"Rv1305\",\n",
-      "      \"entity_type\": \"accession_number\",\n",
-      "      \"char_start\": 114,\n",
-      "      \"char_end\": 120,\n",
-      "      \"attributes\": {},\n",
-      "      \"alignment\": \"match_exact\"\n",
-      "    }\n",
-      "  ],\n",
-      "  \"products\": [],\n",
-      "  \"functional_categories\": [\n",
-      "    {\n",
-      "      \"text\": \"energy metabolism pathway\",\n",
-      "      \"entity_type\": \"functional_category\",\n",
-      "      \"char_start\": 286,\n",
-      "      \"char_end\": 311,\n",
-      "      \"attributes\": {},\n",
-      "      \"alignment\": \"match_exact\"\n",
-      "    }\n",
-      "  ],\n",
-      "  \"screening_methods\": [\n",
-      "    {\n",
-      "      \"text\": \"whole-cell screening\",\n",
-      "      \"entity_type\": \"screening_method\",\n",
-      "      \"char_start\": 249,\n",
-      "      \"char_end\": 269,\n",
-      "      \"attributes\": {},\n",
-      "      \"alignment\": \"match_exact\"\n",
-      "    }\n",
-      "  ],\n",
-      "  \"unclassified\": []\n",
-      "}\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# Serialize to dict (useful for JSON export)\n",
     "import json\n",
@@ -770,33 +695,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "--- Text 1 ---\n",
-      "  compound_name        | Imatinib\n",
-      "  target               | BCR-ABL\n",
-      "  disease              | CML\n",
-      "  bioactivity          | IC50 = 0.6 µM\n",
-      "\n",
-      "--- Text 2 ---\n",
-      "  compound_name        | Trastuzumab\n",
-      "  target               | HER2\n",
-      "  disease              | breast cancer\n",
-      "\n",
-      "--- Text 3 ---\n",
-      "  compound_name        | Remdesivir\n",
-      "  compound_name        | GS-5734\n",
-      "  disease              | SARS-CoV-2\n",
-      "  bioactivity          | EC50 = 0.77 µM\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "texts = [\n",
     "    \"Imatinib inhibits BCR-ABL with IC50 = 0.6 µM in CML.\",\n",
@@ -836,7 +737,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.8"
+   "version": "3.12.2"
   }
  },
  "nbformat": 4,

{structflo_ner-0.3.0 → structflo_ner-0.4.0}/structflo/ner/__init__.py

@@ -63,7 +63,7 @@ from structflo.ner.profiles import (
     EntityProfile,
 )
 
-__version__ = "0.3.0"
+__version__ = "0.4.0"
 
 __all__ = [
     # Main classes

{structflo_ner-0.3.0 → structflo_ner-0.4.0}/structflo/ner/extractor.py

@@ -46,6 +46,12 @@ class NERExtractor:
         model_url: Base URL for self-hosted models (e.g. Ollama at
             ``"http://localhost:11434"``). When set, langextract routes
             requests to this endpoint instead of a cloud API.
+        provider: Explicit langextract provider name (``"ollama"``,
+            ``"openai"``, ``"gemini"``). When set, routing is deterministic and
+            no longer depends on ``model_id`` matching a built-in regex — use
+            this for non-standard model names (e.g. ``"medgemma:latest"``) that
+            would otherwise fail to resolve. When ``None`` (default), the
+            provider is inferred from ``model_id`` as before.
         profile: Default :class:`EntityProfile` to use when no per-call
             profile is specified. Defaults to :data:`FULL`.
         extra_examples: Additional :class:`lx.data.ExampleData` objects that
@@ -63,6 +69,7 @@ class NERExtractor:
         model_id: str = "gemini-2.5-flash",
         api_key: str | None = None,
         model_url: str | None = None,
+        provider: str | None = None,
         profile: EntityProfile = FULL,
         extra_examples: list[lx.data.ExampleData] | None = None,
         langextract_kwargs: dict | None = None,
@@ -70,6 +77,7 @@ class NERExtractor:
         self._model_id = model_id
         self._api_key = api_key
         self._model_url = model_url
+        self._provider = provider
         self._default_profile = profile
         self._extra_examples = extra_examples or []
         self._langextract_kwargs = langextract_kwargs or {}
@@ -159,7 +167,30 @@ class NERExtractor:
     @property
     def _is_ollama(self) -> bool:
         """Return True when routing to an Ollama endpoint."""
-        return self._model_url is not None
+        return self._provider == "ollama" or (self._provider is None and self._model_url is not None)
+
+    def _build_config(self) -> "lx.factory.ModelConfig":
+        """Build an explicit ModelConfig for deterministic provider routing.
+
+        Provider-specific settings travel in ``provider_kwargs`` (passed to the
+        provider constructor): ``api_key`` for cloud providers, ``base_url`` +
+        ``num_ctx`` for Ollama.
+        """
+        from langextract import factory
+
+        provider_kwargs: dict = {}
+        if self._api_key is not None:
+            provider_kwargs["api_key"] = self._api_key
+        if self._model_url is not None:
+            provider_kwargs["base_url"] = self._model_url
+        if self._is_ollama:
+            # Ollama defaults to num_ctx=2048, far too small for few-shot NER.
+            provider_kwargs.setdefault("num_ctx", 8192)
+        return factory.ModelConfig(
+            model_id=self._model_id,
+            provider=self._provider,
+            provider_kwargs=provider_kwargs,
+        )
 
     def _run_extraction(
         self,
@@ -174,22 +205,29 @@ class NERExtractor:
         kwargs.setdefault("use_schema_constraints", True)
         kwargs.setdefault("show_progress", False)
 
-        # Ollama defaults to num_ctx=2048 which is far too small for
-        # few-shot NER prompts.  Set a sane default so users don't hit
-        # silent truncation.
-        if self._is_ollama:
-            lm_params = kwargs.setdefault("language_model_params", {})
-            lm_params.setdefault("num_ctx", 8192)
-
-        result = lx.extract(
-            text_or_documents=text,
-            prompt_description=prompt,
-            examples=examples,
-            model_id=self._model_id,
-            api_key=self._api_key,
-            model_url=self._model_url,
-            **kwargs,
-        )
+        if self._provider is not None:
+            # Explicit provider → deterministic routing via ModelConfig.
+            result = lx.extract(
+                text_or_documents=text,
+                prompt_description=prompt,
+                examples=examples,
+                config=self._build_config(),
+                **kwargs,
+            )
+        else:
+            # Legacy path: provider inferred from model_id regex by langextract.
+            if self._is_ollama:
+                lm_params = kwargs.setdefault("language_model_params", {})
+                lm_params.setdefault("num_ctx", 8192)
+            result = lx.extract(
+                text_or_documents=text,
+                prompt_description=prompt,
+                examples=examples,
+                model_id=self._model_id,
+                api_key=self._api_key,
+                model_url=self._model_url,
+                **kwargs,
+            )
 
         # lx.extract returns a list when given a list; we always pass a single string
         doc = result[0] if isinstance(result, list) else result

{structflo_ner-0.3.0 → structflo_ner-0.4.0}/tests/test_extractor.py

@@ -122,6 +122,74 @@ class TestNERExtractorExtract:
         assert result.source_text == "My source text"
 
 
+class TestProviderRouting:
+    """Explicit provider selection routes deterministically via ModelConfig,
+    instead of relying on langextract's model_id regex inference."""
+
+    def _run_and_capture(self, monkeypatch, **init_kwargs):
+        from structflo.ner import extractor as extractor_mod
+
+        captured = {}
+
+        def fake_extract(**kwargs):
+            captured.update(kwargs)
+            return _make_annotated_doc([])
+
+        monkeypatch.setattr(extractor_mod.lx, "extract", fake_extract)
+        NERExtractor(**init_kwargs).extract("some text")
+        return captured
+
+    def test_provider_defaults_to_none(self):
+        assert NERExtractor()._provider is None
+
+    def test_provider_stored(self):
+        assert NERExtractor(provider="openai")._provider == "openai"
+
+    def test_no_provider_uses_legacy_model_id_path(self, monkeypatch):
+        captured = self._run_and_capture(
+            monkeypatch, model_id="gemma3:27b", model_url="http://ollama:11434"
+        )
+        assert captured["model_id"] == "gemma3:27b"
+        assert captured["model_url"] == "http://ollama:11434"
+        assert "config" not in captured
+
+    def test_explicit_provider_uses_config_not_top_level_model_id(self, monkeypatch):
+        captured = self._run_and_capture(
+            monkeypatch,
+            provider="ollama",
+            model_id="medgemma:latest",
+            model_url="http://ollama:11434",
+        )
+        # Deterministic routing: config carries provider + model_id, and the
+        # top-level model_id/model_url are NOT passed (would conflict with config).
+        assert "config" in captured
+        assert "model_id" not in captured
+        assert "model_url" not in captured
+        config = captured["config"]
+        assert config.provider == "ollama"
+        assert config.model_id == "medgemma:latest"
+
+    def test_ollama_provider_kwargs_carry_base_url_and_num_ctx(self, monkeypatch):
+        captured = self._run_and_capture(
+            monkeypatch,
+            provider="ollama",
+            model_id="medgemma:latest",
+            model_url="http://ollama:11434",
+        )
+        pk = captured["config"].provider_kwargs
+        assert pk["base_url"] == "http://ollama:11434"
+        assert pk["num_ctx"] == 8192
+
+    def test_cloud_provider_kwargs_carry_api_key(self, monkeypatch):
+        captured = self._run_and_capture(
+            monkeypatch, provider="openai", model_id="gpt-4o", api_key="sk-test"
+        )
+        pk = captured["config"].provider_kwargs
+        assert pk["api_key"] == "sk-test"
+        assert "num_ctx" not in pk
+        assert "base_url" not in pk
+
+
 class TestBuildPrompt:
     def test_prompt_includes_entity_class_constraint(self):
         extractor = NERExtractor()