Gene expression changes in aging mouse brain predicting neurodegenerative vulnerability¶
Notebook ID: nb-SDA-2026-04-03-gap-aging-mouse-brain-v3-20260402 · Analysis: SDA-2026-04-03-gap-aging-mouse-brain-v3-20260402 · Generated: 2026-04-20T09:01:23
Research question¶
What gene expression changes in the aging mouse brain predict neurodegenerative vulnerability? Use Allen Aging Mouse Brain Atlas data. Cross-reference with human AD datasets. Produce hypotheses about aging-neurodegeneration mechanisms.
Approach¶
This notebook is generated programmatically from real Forge tool calls and SciDEX debate data. Forge tools used: PubMed Search, MyGene, STRING PPI, Reactome pathways, Enrichr.
Debate Summary¶
Quality score: 0.95 · Rounds: 4
1. Target gene annotations (MyGene + Human Protein Atlas)¶
import pandas as pd
ann_rows = [{'gene': 'ACE', 'name': 'angiotensin I converting enzyme', 'protein_class': "['Candidate cardiovascular disease genes', 'CD markers', 'Di", 'disease_involvement': "['FDA approved drug targets']"}, {'gene': 'AND', 'name': '—', 'protein_class': '—', 'disease_involvement': '—'}, {'gene': 'AP1S1', 'name': 'adaptor related protein complex 1 subunit sigma 1', 'protein_class': "['Disease related genes', 'Human disease related genes', 'Pl", 'disease_involvement': "['Deafness', 'Ichthyosis', 'Intellectual disability', 'Neuropathy']"}, {'gene': 'APOE', 'name': 'apolipoprotein E', 'protein_class': "['Cancer-related genes', 'Candidate cardiovascular disease g", 'disease_involvement': "['Alzheimer disease', 'Amyloidosis', 'Cancer-related genes', 'Disease variant', "}, {'gene': 'APP', 'name': 'amyloid beta precursor protein', 'protein_class': "['Disease related genes', 'FDA approved drug targets', 'Huma", 'disease_involvement': "['Alzheimer disease', 'Amyloidosis', 'Disease variant', 'FDA approved drug targe"}, {'gene': 'C1QA', 'name': 'complement C1q A chain', 'protein_class': "['Disease related genes', 'Human disease related genes', 'Pl", 'disease_involvement': '—'}, {'gene': 'C4B', 'name': 'complement C4B (Chido/Rodgers blood group)', 'protein_class': '—', 'disease_involvement': '—'}, {'gene': 'CD300F', 'name': 'CD300 molecule like family member f', 'protein_class': '—', 'disease_involvement': '—'}, {'gene': 'CDKN2A', 'name': 'cyclin dependent kinase inhibitor 2A', 'protein_class': "['Cancer-related genes', 'Disease related genes', 'Human dis", 'disease_involvement': "['Cancer-related genes', 'Disease variant', 'Li-Fraumeni syndrome', 'Tumor suppr"}, {'gene': 'CGAS', 'name': 'cyclic GMP-AMP synthase', 'protein_class': "['Enzymes', 'Predicted intracellular proteins']", 'disease_involvement': '—'}, {'gene': 'COMPLEXES', 'name': 'VPS16 core subunit of CORVET and HOPS complexes', 'protein_class': '—', 'disease_involvement': '—'}, {'gene': 'CSF1R', 'name': 'colony stimulating factor 1 receptor', 'protein_class': "['Cancer-related genes', 'CD markers', 'Disease related gene", 'disease_involvement': "['Cancer-related genes', 'Disease variant', 'FDA approved drug targets', 'Neurod"}, {'gene': 'CXCL10', 'name': 'C-X-C motif chemokine ligand 10', 'protein_class': "['Cancer-related genes', 'Human disease related genes', 'Pla", 'disease_involvement': "['Cancer-related genes']"}, {'gene': 'CYP46A1', 'name': 'cytochrome P450 family 46 subfamily A member 1', 'protein_class': "['Enzymes', 'Metabolic proteins', 'Predicted intracellular p", 'disease_involvement': '—'}, {'gene': 'CYTOKINE', 'name': 'IK cytokine', 'protein_class': '—', 'disease_involvement': '—'}, {'gene': 'GAL3ST1', 'name': 'galactose-3-O-sulfotransferase 1', 'protein_class': "['Enzymes', 'Metabolic proteins', 'Predicted intracellular p", 'disease_involvement': '—'}, {'gene': 'GPX4', 'name': 'glutathione peroxidase 4', 'protein_class': "['Disease related genes', 'Enzymes', 'Essential proteins', '", 'disease_involvement': "['Dwarfism']"}, {'gene': 'INFLAMMATORY', 'name': 'Inflammatory bowel disease 12', 'protein_class': '—', 'disease_involvement': '—'}, {'gene': 'MARKERS', 'name': 'deafness, autosomal dominant 58', 'protein_class': '—', 'disease_involvement': '—'}, {'gene': 'MITOCHONDRIAL', 'name': 'ferritin mitochondrial', 'protein_class': '—', 'disease_involvement': '—'}]
pd.DataFrame(ann_rows)
| gene | name | protein_class | disease_involvement | |
|---|---|---|---|---|
| 0 | ACE | angiotensin I converting enzyme | ['Candidate cardiovascular disease genes', 'CD... | ['FDA approved drug targets'] |
| 1 | AND | — | — | — |
| 2 | AP1S1 | adaptor related protein complex 1 subunit sigma 1 | ['Disease related genes', 'Human disease relat... | ['Deafness', 'Ichthyosis', 'Intellectual disab... |
| 3 | APOE | apolipoprotein E | ['Cancer-related genes', 'Candidate cardiovasc... | ['Alzheimer disease', 'Amyloidosis', 'Cancer-r... |
| 4 | APP | amyloid beta precursor protein | ['Disease related genes', 'FDA approved drug t... | ['Alzheimer disease', 'Amyloidosis', 'Disease ... |
| 5 | C1QA | complement C1q A chain | ['Disease related genes', 'Human disease relat... | — |
| 6 | C4B | complement C4B (Chido/Rodgers blood group) | — | — |
| 7 | CD300F | CD300 molecule like family member f | — | — |
| 8 | CDKN2A | cyclin dependent kinase inhibitor 2A | ['Cancer-related genes', 'Disease related gene... | ['Cancer-related genes', 'Disease variant', 'L... |
| 9 | CGAS | cyclic GMP-AMP synthase | ['Enzymes', 'Predicted intracellular proteins'] | — |
| 10 | COMPLEXES | VPS16 core subunit of CORVET and HOPS complexes | — | — |
| 11 | CSF1R | colony stimulating factor 1 receptor | ['Cancer-related genes', 'CD markers', 'Diseas... | ['Cancer-related genes', 'Disease variant', 'F... |
| 12 | CXCL10 | C-X-C motif chemokine ligand 10 | ['Cancer-related genes', 'Human disease relate... | ['Cancer-related genes'] |
| 13 | CYP46A1 | cytochrome P450 family 46 subfamily A member 1 | ['Enzymes', 'Metabolic proteins', 'Predicted i... | — |
| 14 | CYTOKINE | IK cytokine | — | — |
| 15 | GAL3ST1 | galactose-3-O-sulfotransferase 1 | ['Enzymes', 'Metabolic proteins', 'Predicted i... | — |
| 16 | GPX4 | glutathione peroxidase 4 | ['Disease related genes', 'Enzymes', 'Essentia... | ['Dwarfism'] |
| 17 | INFLAMMATORY | Inflammatory bowel disease 12 | — | — |
| 18 | MARKERS | deafness, autosomal dominant 58 | — | — |
| 19 | MITOCHONDRIAL | ferritin mitochondrial | — | — |
2. GO Biological Process enrichment (Enrichr)¶
go_bp = [{'rank': 1, 'term': 'Regulation Of Long-Term Synaptic Potentiation (GO:1900271)', 'p_value': 3.775486561735298e-06, 'odds_ratio': 125.74789915966386, 'genes': ['APP', 'APOE', 'CYP46A1']}, {'rank': 2, 'term': 'Regulation Of Protein Tyrosine Kinase Activity (GO:0061097)', 'p_value': 7.637120838980064e-06, 'odds_ratio': 97.76470588235294, 'genes': ['APP', 'CSF1R', 'ACE']}, {'rank': 3, 'term': 'Sterol Catabolic Process (GO:0016127)', 'p_value': 1.4215693891863761e-05, 'odds_ratio': 554.8888888888889, 'genes': ['APOE', 'CYP46A1']}, {'rank': 4, 'term': 'Cholesterol Catabolic Process (GO:0006707)', 'p_value': 1.4215693891863761e-05, 'odds_ratio': 554.8888888888889, 'genes': ['APOE', 'CYP46A1']}, {'rank': 5, 'term': 'Negative Regulation Of Long-Term Synaptic Potentiation (GO:1900272)', 'p_value': 2.650430025143172e-05, 'odds_ratio': 369.8888888888889, 'genes': ['APP', 'APOE']}, {'rank': 6, 'term': 'Alcohol Catabolic Process (GO:0046164)', 'p_value': 2.650430025143172e-05, 'odds_ratio': 369.8888888888889, 'genes': ['APOE', 'CYP46A1']}, {'rank': 7, 'term': 'Neuron Remodeling (GO:0016322)', 'p_value': 3.405659800132413e-05, 'odds_ratio': 317.031746031746, 'genes': ['APP', 'C1QA']}, {'rank': 8, 'term': 'Astrocyte Activation (GO:0048143)', 'p_value': 6.232515501994467e-05, 'odds_ratio': 221.88888888888889, 'genes': ['APP', 'C1QA']}, {'rank': 9, 'term': 'Astrocyte Development (GO:0014002)', 'p_value': 9.897575036981038e-05, 'odds_ratio': 170.65811965811966, 'genes': ['APP', 'C1QA']}, {'rank': 10, 'term': 'Positive Regulation Of Lymphocyte Migration (GO:2000403)', 'p_value': 0.00011304743026189121, 'odds_ratio': 158.46031746031747, 'genes': ['APP', 'CXCL10']}]
go_df = pd.DataFrame(go_bp)[['term','p_value','odds_ratio','genes']]
go_df['p_value'] = go_df['p_value'].apply(lambda p: f'{p:.2e}')
go_df['odds_ratio'] = go_df['odds_ratio'].round(1)
go_df['term'] = go_df['term'].str[:60]
go_df['n_hits'] = go_df['genes'].apply(len)
go_df['genes'] = go_df['genes'].apply(lambda g: ', '.join(g))
go_df[['term','n_hits','p_value','odds_ratio','genes']]
| term | n_hits | p_value | odds_ratio | genes | |
|---|---|---|---|---|---|
| 0 | Regulation Of Long-Term Synaptic Potentiation ... | 3 | 3.78e-06 | 125.7 | APP, APOE, CYP46A1 |
| 1 | Regulation Of Protein Tyrosine Kinase Activity... | 3 | 7.64e-06 | 97.8 | APP, CSF1R, ACE |
| 2 | Sterol Catabolic Process (GO:0016127) | 2 | 1.42e-05 | 554.9 | APOE, CYP46A1 |
| 3 | Cholesterol Catabolic Process (GO:0006707) | 2 | 1.42e-05 | 554.9 | APOE, CYP46A1 |
| 4 | Negative Regulation Of Long-Term Synaptic Pote... | 2 | 2.65e-05 | 369.9 | APP, APOE |
| 5 | Alcohol Catabolic Process (GO:0046164) | 2 | 2.65e-05 | 369.9 | APOE, CYP46A1 |
| 6 | Neuron Remodeling (GO:0016322) | 2 | 3.41e-05 | 317.0 | APP, C1QA |
| 7 | Astrocyte Activation (GO:0048143) | 2 | 6.23e-05 | 221.9 | APP, C1QA |
| 8 | Astrocyte Development (GO:0014002) | 2 | 9.90e-05 | 170.7 | APP, C1QA |
| 9 | Positive Regulation Of Lymphocyte Migration (G... | 2 | 1.13e-04 | 158.5 | APP, CXCL10 |
import matplotlib.pyplot as plt
import numpy as np
go_bp = [{'rank': 1, 'term': 'Regulation Of Long-Term Synaptic Potentiation (GO:1900271)', 'p_value': 3.775486561735298e-06, 'odds_ratio': 125.74789915966386, 'genes': ['APP', 'APOE', 'CYP46A1']}, {'rank': 2, 'term': 'Regulation Of Protein Tyrosine Kinase Activity (GO:0061097)', 'p_value': 7.637120838980064e-06, 'odds_ratio': 97.76470588235294, 'genes': ['APP', 'CSF1R', 'ACE']}, {'rank': 3, 'term': 'Sterol Catabolic Process (GO:0016127)', 'p_value': 1.4215693891863761e-05, 'odds_ratio': 554.8888888888889, 'genes': ['APOE', 'CYP46A1']}, {'rank': 4, 'term': 'Cholesterol Catabolic Process (GO:0006707)', 'p_value': 1.4215693891863761e-05, 'odds_ratio': 554.8888888888889, 'genes': ['APOE', 'CYP46A1']}, {'rank': 5, 'term': 'Negative Regulation Of Long-Term Synaptic Potentiation (GO:1900272)', 'p_value': 2.650430025143172e-05, 'odds_ratio': 369.8888888888889, 'genes': ['APP', 'APOE']}, {'rank': 6, 'term': 'Alcohol Catabolic Process (GO:0046164)', 'p_value': 2.650430025143172e-05, 'odds_ratio': 369.8888888888889, 'genes': ['APOE', 'CYP46A1']}, {'rank': 7, 'term': 'Neuron Remodeling (GO:0016322)', 'p_value': 3.405659800132413e-05, 'odds_ratio': 317.031746031746, 'genes': ['APP', 'C1QA']}, {'rank': 8, 'term': 'Astrocyte Activation (GO:0048143)', 'p_value': 6.232515501994467e-05, 'odds_ratio': 221.88888888888889, 'genes': ['APP', 'C1QA']}]
terms = [t['term'][:45] for t in go_bp][::-1]
neglogp = [-np.log10(max(t['p_value'], 1e-300)) for t in go_bp][::-1]
fig, ax = plt.subplots(figsize=(9, 4.5))
ax.barh(terms, neglogp, color='#4fc3f7')
ax.set_xlabel('-log10(p-value)')
ax.set_title('Top GO:BP enrichment (Enrichr)')
ax.grid(axis='x', alpha=0.3)
plt.tight_layout(); plt.show()
3. STRING protein interaction network¶
ppi = [{'protein1': 'APOE', 'protein2': 'APP', 'score': 0.995, 'nscore': 0, 'fscore': 0, 'pscore': 0, 'ascore': 0, 'escore': 0.621, 'dscore': 0.72, 'tscore': 0.96}, {'protein1': 'CSF1R', 'protein2': 'C1QA', 'score': 0.456, 'nscore': 0, 'fscore': 0, 'pscore': 0, 'ascore': 0, 'escore': 0, 'dscore': 0, 'tscore': 0.456}]
ppi_df = pd.DataFrame(ppi).sort_values('score', ascending=False)
display_cols = [c for c in ['protein1','protein2','score','escore','tscore'] if c in ppi_df.columns]
print(f'{len(ppi_df)} STRING edges')
ppi_df[display_cols].head(20)
2 STRING edges
| protein1 | protein2 | score | escore | tscore | |
|---|---|---|---|---|---|
| 0 | APOE | APP | 0.995 | 0.621 | 0.960 |
| 1 | CSF1R | C1QA | 0.456 | 0.000 | 0.456 |
import math
ppi = [{'protein1': 'APOE', 'protein2': 'APP', 'score': 0.995, 'nscore': 0, 'fscore': 0, 'pscore': 0, 'ascore': 0, 'escore': 0.621, 'dscore': 0.72, 'tscore': 0.96}, {'protein1': 'CSF1R', 'protein2': 'C1QA', 'score': 0.456, 'nscore': 0, 'fscore': 0, 'pscore': 0, 'ascore': 0, 'escore': 0, 'dscore': 0, 'tscore': 0.456}]
if ppi:
nodes = sorted({p for e in ppi for p in (e['protein1'], e['protein2'])})
n = len(nodes)
pos = {n_: (math.cos(2*math.pi*i/n), math.sin(2*math.pi*i/n)) for i, n_ in enumerate(nodes)}
fig, ax = plt.subplots(figsize=(7, 7))
for e in ppi:
x1,y1 = pos[e['protein1']]; x2,y2 = pos[e['protein2']]
ax.plot([x1,x2],[y1,y2], color='#888', alpha=0.3+0.5*e.get('score',0))
for name,(x,y) in pos.items():
ax.scatter([x],[y], s=450, color='#ffd54f', edgecolors='#333', zorder=3)
ax.annotate(name, (x,y), ha='center', va='center', fontsize=8, fontweight='bold', zorder=4)
ax.set_aspect('equal'); ax.axis('off')
ax.set_title(f'STRING PPI network ({len(ppi)} edges)')
plt.tight_layout(); plt.show()
4. Reactome pathway footprint¶
pw_rows = [{'gene': 'ACE', 'n_pathways': 1, 'top_pathway': 'Metabolism of Angiotensinogen to Angiotensins'}, {'gene': 'AND', 'n_pathways': 0, 'top_pathway': '—'}, {'gene': 'AP1S1', 'n_pathways': 4, 'top_pathway': 'Nef mediated downregulation of MHC class I complex cell surface expres'}, {'gene': 'APOE', 'n_pathways': 8, 'top_pathway': 'Nuclear signaling by ERBB4'}, {'gene': 'APP', 'n_pathways': 8, 'top_pathway': 'Platelet degranulation'}, {'gene': 'C1QA', 'n_pathways': 3, 'top_pathway': 'Initial triggering of complement'}, {'gene': 'C4B', 'n_pathways': 3, 'top_pathway': 'Initial triggering of complement'}, {'gene': 'CD300F', 'n_pathways': 1, 'top_pathway': 'Immunoregulatory interactions between a Lymphoid and a non-Lymphoid ce'}, {'gene': 'CDKN2A', 'n_pathways': 0, 'top_pathway': '—'}, {'gene': 'CGAS', 'n_pathways': 1, 'top_pathway': 'STING mediated induction of host immune responses'}, {'gene': 'COMPLEXES', 'n_pathways': 0, 'top_pathway': '—'}, {'gene': 'CSF1R', 'n_pathways': 3, 'top_pathway': 'Other interleukin signaling'}, {'gene': 'CXCL10', 'n_pathways': 3, 'top_pathway': 'Chemokine receptors bind chemokines'}, {'gene': 'CYP46A1', 'n_pathways': 2, 'top_pathway': 'Synthesis of bile acids and bile salts via 24-hydroxycholesterol'}, {'gene': 'CYTOKINE', 'n_pathways': 0, 'top_pathway': '—'}, {'gene': 'GAL3ST1', 'n_pathways': 1, 'top_pathway': 'Glycosphingolipid biosynthesis'}, {'gene': 'GPX4', 'n_pathways': 7, 'top_pathway': 'Synthesis of 5-eicosatetraenoic acids'}, {'gene': 'INFLAMMATORY', 'n_pathways': 0, 'top_pathway': '—'}, {'gene': 'MARKERS', 'n_pathways': 0, 'top_pathway': '—'}, {'gene': 'MITOCHONDRIAL', 'n_pathways': 0, 'top_pathway': '—'}]
pd.DataFrame(pw_rows).sort_values('n_pathways', ascending=False)
| gene | n_pathways | top_pathway | |
|---|---|---|---|
| 3 | APOE | 8 | Nuclear signaling by ERBB4 |
| 4 | APP | 8 | Platelet degranulation |
| 16 | GPX4 | 7 | Synthesis of 5-eicosatetraenoic acids |
| 2 | AP1S1 | 4 | Nef mediated downregulation of MHC class I com... |
| 12 | CXCL10 | 3 | Chemokine receptors bind chemokines |
| 11 | CSF1R | 3 | Other interleukin signaling |
| 6 | C4B | 3 | Initial triggering of complement |
| 5 | C1QA | 3 | Initial triggering of complement |
| 13 | CYP46A1 | 2 | Synthesis of bile acids and bile salts via 24-... |
| 0 | ACE | 1 | Metabolism of Angiotensinogen to Angiotensins |
| 9 | CGAS | 1 | STING mediated induction of host immune responses |
| 7 | CD300F | 1 | Immunoregulatory interactions between a Lympho... |
| 15 | GAL3ST1 | 1 | Glycosphingolipid biosynthesis |
| 1 | AND | 0 | — |
| 10 | COMPLEXES | 0 | — |
| 8 | CDKN2A | 0 | — |
| 14 | CYTOKINE | 0 | — |
| 17 | INFLAMMATORY | 0 | — |
| 18 | MARKERS | 0 | — |
| 19 | MITOCHONDRIAL | 0 | — |
5. Hypothesis ranking (39 hypotheses)¶
hyp_data = [('APOE-TREM2 Ligand Availability Dysfunction in Neurodege', 0), ('TREM2-Driven Senescence Biomarker Index for Predicting ', 0), ('TREM2-Mediated Mitochondrial Dysfunction in Neurodegene', 0), ('Age-Dependent TREM2 Signaling Disrupts Astrocyte-Microg', 0), ('TREM2-Mediated Oligodendrocyte-Microglia Metabolic Coup', 0), ('TREM2-Dependent Astrocyte-Microglia Cross-talk in Neuro', 0.99), ('TREM2-Dependent Microglial Senescence Transition', 0.95), ('TREM2-ASM Crosstalk in Microglial Lysosomal Senescence', 0.91), ('TREM2-Mediated Astrocyte-Microglia Cross-Talk in Neurod', 0.902), ('SIRT1-Mediated Reversal of TREM2-Dependent Microglial S', 0.895), ('TREM2-Mediated Astrocyte-Microglia Crosstalk in Neurode', 0.892), ('TREM2-CSF1R Cross-Talk in Microglial Metabolic Reprogra', 0.885), ('TREM2-SIRT1 Metabolic Senescence Circuit in Microglial ', 0.882), ('TREM2-Mediated Astrocyte-Microglia Cross-Talk in Neurod', 0.88), ('TREM2-Mediated Astrocyte-Microglia Cross-Talk in Neurod', 0.875), ('TREM2-Mediated Cholesterol Dysregulation in Microglial ', 0.869), ('Early Proteasome Restoration Therapy', 0.712), ('Ferroptosis Inhibition for α-Synuclein Neuroprotection', 0.705), ('cGAS-STING Senescence Circuit Disruption', 0.691), ('White Matter Oligodendrocyte Protection via CXCL10 Inhi', 0.675), ('Age-Dependent Complement C4b Upregulation Drives Synapt', 0.671), ('White Matter Vulnerability Prevention via Oligodendrocy', 0.667), ('Oligodendrocyte White Matter Vulnerability', 0.651), ('Oligodendrocyte Remyelination Enhancement', 0.644), ('White Matter Immune Checkpoint Restoration', 0.644), ('Mitochondrial NAD+ Salvage Enhancement', 0.639), ('Selective Neuronal Vulnerability Network Targeting', 0.638), ('Selective Cholinergic Protection via APP Pathway Modula', 0.629), ('Myelin Sulfatide Restoration', 0.623), ('TFEB-PGC1α Mitochondrial-Lysosomal Decoupling', 0.622), ('Microglial ACE Enhancement for Amyloid Clearance', 0.622), ('Mitochondrial-Cytokine Axis Modulation', 0.616), ('Complement-Mediated Synaptic Pruning Dysregulation', 0.612), ('AP1S1-Mediated Vesicular Transport Restoration', 0.588), ('TNFRSF25-Mediated Aging Exosome Pathway Inhibition', 0.587), ('Senescence-Tau Decoupling Therapy', 0.585), ('NOMO1-Mediated Neuronal Resilience Enhancement', 0.584), ('Profilin-1 Cytoskeletal Checkpoint Enhancement', 0.554), ('CD300f Immune Checkpoint Activation', 0.545)]
titles = [h[0] for h in hyp_data][::-1]
scores = [h[1] for h in hyp_data][::-1]
fig, ax = plt.subplots(figsize=(10, max(8, len(titles)*0.4)))
colors = ['#ef5350' if s >= 0.6 else '#ffa726' if s >= 0.5 else '#66bb6a' for s in scores]
ax.barh(range(len(titles)), scores, color=colors)
ax.set_yticks(range(len(titles))); ax.set_yticklabels(titles, fontsize=7)
ax.set_xlabel('Composite Score'); ax.set_title('Gene expression changes in aging mouse brain predicting neurodegenerative vulnerability')
ax.grid(axis='x', alpha=0.3)
plt.tight_layout(); plt.show()
6. Score dimension heatmap (top 10)¶
labels = ['APOE-TREM2 Ligand Availability Dysfuncti', 'TREM2-Driven Senescence Biomarker Index ', 'TREM2-Mediated Mitochondrial Dysfunction', 'Age-Dependent TREM2 Signaling Disrupts A', 'TREM2-Mediated Oligodendrocyte-Microglia', 'TREM2-Dependent Astrocyte-Microglia Cros', 'TREM2-Dependent Microglial Senescence Tr', 'TREM2-ASM Crosstalk in Microglial Lysoso', 'TREM2-Mediated Astrocyte-Microglia Cross', 'SIRT1-Mediated Reversal of TREM2-Depende']
matrix = np.array([[0.72, 0.82, 0.78, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58], [0.7, 0.8, 0.76, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58], [0.72, 0.82, 0.78, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58], [0.72, 0.82, 0.78, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58], [0.72, 0.82, 0.78, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58], [0.72, 0.82, 0.78, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58], [0.78, 0.72, 0.91, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58], [0.7, 0.8, 0.76, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58], [0.72, 0.82, 0.78, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58], [0.7, 0.8, 0.76, 0.88, 0.263, 0.85, 0.75, 0.65, 0.58]])
dims = ['novelty_score', 'feasibility_score', 'impact_score', 'mechanistic_plausibility_score', 'clinical_relevance_score', 'data_availability_score', 'reproducibility_score', 'druggability_score', 'safety_profile_score']
if matrix.size:
fig, ax = plt.subplots(figsize=(10, 5))
im = ax.imshow(matrix, cmap='RdYlGn', aspect='auto', vmin=0, vmax=1)
ax.set_xticks(range(len(dims)))
ax.set_xticklabels([d.replace('_score','').replace('_',' ').title() for d in dims],
rotation=45, ha='right', fontsize=8)
ax.set_yticks(range(len(labels))); ax.set_yticklabels(labels, fontsize=7)
ax.set_title('Score dimensions — top hypotheses')
plt.colorbar(im, ax=ax, shrink=0.8)
plt.tight_layout(); plt.show()
else:
print('No score data available')
7. PubMed literature per hypothesis¶
Hypothesis 1: APOE-TREM2 Ligand Availability Dysfunction in Neurodegeneration¶
Target genes: APOE · Composite score: 0.0
Background and Rationale
While TREM2 dysfunction has been implicated in neurodegeneration, the primary driver may not be age-related senescence transitions but rather disrupted ligand availability and recognition. TREM2's neuroprotective functions depend critically on its interaction with endog
print('No PubMed results for hypothesis h-var-9d0c0787a5')
No PubMed results for hypothesis h-var-9d0c0787a5
Hypothesis 2: TREM2-Driven Senescence Biomarker Index for Predicting Neurodegenerati¶
Target genes: TREM2 · Composite score: 0.0
This hypothesis proposes that the TREM2-dependent microglial senescence transition can be tracked and predicted through a novel composite biomarker index combining plasma p-tau217/CSF neurogranin ratio with microglial senescence markers. As TREM2 signaling undergoes age-related dysfunction, transiti
print('No PubMed results for hypothesis h-var-56a6156e67')
No PubMed results for hypothesis h-var-56a6156e67
Hypothesis 3: TREM2-Mediated Mitochondrial Dysfunction in Neurodegeneration¶
Target genes: TREM2 · Composite score: 0.0
Background and Rationale
TREM2 loss-of-function variants significantly increase Alzheimer's disease risk through mechanisms beyond traditional inflammatory pathways. While TREM2's role in microglial activation is well-established, emerging evidence suggests that TREM2 signaling directly regulat
print('No PubMed results for hypothesis h-var-a8c6c6ea92')
No PubMed results for hypothesis h-var-a8c6c6ea92
Hypothesis 4: Age-Dependent TREM2 Signaling Disrupts Astrocyte-Microglia Communicati¶
Target genes: TREM2 · Composite score: 0.0
This hypothesis proposes that age-related changes in TREM2 signaling create a pathological cascade where senescent microglia progressively lose their ability to maintain protective astrocyte-microglia communication networks. Under physiological conditions, TREM2 engagement triggers microglial releas
print('No PubMed results for hypothesis h-var-71ac892791')
No PubMed results for hypothesis h-var-71ac892791
Hypothesis 5: TREM2-Mediated Oligodendrocyte-Microglia Metabolic Coupling in White M¶
Target genes: TREM2 · Composite score: 0.0
Background and Rationale TREM2 variants are established risk factors for Alzheimer's disease, but emerging evidence indicates significant white matter pathology precedes cortical neurodegeneration. While TREM2 is exclusively expressed on microglia, recent studies reveal that TREM2+ microglia are
print('No PubMed results for hypothesis h-var-33f643b43a')
No PubMed results for hypothesis h-var-33f643b43a
Hypothesis 6: TREM2-Dependent Astrocyte-Microglia Cross-talk in Neurodegeneration¶
Target genes: TREM2 · Composite score: 0.99
Background and Rationale TREM2 variants represent major genetic risk factors for Alzheimer's disease, with loss-of-function mutations increasing dementia risk threefold. While TREM2 is exclusively expressed on microglia, emerging evidence suggests its primary pathogenic role occurs through disru
print('No PubMed results for hypothesis h-var-de1677a080')
No PubMed results for hypothesis h-var-de1677a080
Hypothesis 7: TREM2-Dependent Microglial Senescence Transition¶
Target genes: TREM2 · Composite score: 0.95
Background and Rationale
Triggering Receptor Expressed on Myeloid cells 2 (TREM2) represents one of the most significant genetic risk factors for late-onset Alzheimer's disease, with rare loss-of-function variants conferring up to threefold increased risk of dementia. This single-pass transmemb
print('No PubMed results for hypothesis h-61196ade')
No PubMed results for hypothesis h-61196ade
Hypothesis 8: TREM2-ASM Crosstalk in Microglial Lysosomal Senescence¶
Target genes: SMPD1 · Composite score: 0.91
This hypothesis proposes that TREM2-dependent microglial senescence is mechanistically driven by dysregulated acid sphingomyelinase (ASM) activity and ceramide accumulation in aging microglia. During normal aging, TREM2 signaling undergoes fundamental changes that disrupt sphingolipid homeostasis, l
print('No PubMed results for hypothesis h-var-18aae53fe9')
No PubMed results for hypothesis h-var-18aae53fe9
Hypothesis 9: TREM2-Mediated Astrocyte-Microglia Cross-Talk in Neurodegeneration¶
Target genes: TREM2 · Composite score: 0.902
Background and Rationale
Triggering Receptor Expressed on Myeloid cells 2 (TREM2) represents one of the most significant genetic risk factors for late-onset Alzheimer's disease, with rare loss-of-function variants conferring up to threefold increased risk of dementia. While TREM2 is exclusively
print('No PubMed results for hypothesis h-var-bed9f3b7ed')
No PubMed results for hypothesis h-var-bed9f3b7ed
Hypothesis 10: SIRT1-Mediated Reversal of TREM2-Dependent Microglial Senescence¶
Target genes: SIRT1 · Composite score: 0.895
This hypothesis proposes that targeted epigenetic reactivation of SIRT1 can reverse the age-related senescence transition in microglia that is mediated by TREM2 signaling dysfunction. During aging, TREM2-expressing microglia undergo a pathological transition from neuroprotective to neurotoxic phenot
print('No PubMed results for hypothesis h-var-b7de826706')
No PubMed results for hypothesis h-var-b7de826706
Hypothesis 11: TREM2-Mediated Astrocyte-Microglia Crosstalk in Neurodegeneration¶
Target genes: TREM2 · Composite score: 0.892
Background and Rationale
While TREM2 is exclusively expressed on microglia, emerging evidence suggests that TREM2-dependent microglial dysfunction fundamentally disrupts astrocyte-microglia communication networks, creating a pathological feedback loop that accelerates neurodegeneration. Astrocy
print('No PubMed results for hypothesis h-var-66156774e7')
No PubMed results for hypothesis h-var-66156774e7
Hypothesis 12: TREM2-CSF1R Cross-Talk in Microglial Metabolic Reprogramming¶
Target genes: TREM2, CSF1R · Composite score: 0.885
Background and Rationale
TREM2 loss-of-function variants confer significant risk for late-onset Alzheimer's disease, but the mechanisms linking TREM2 dysfunction to neurodegeneration remain incompletely understood. Recent evidence suggests that TREM2 signaling intersects with colony-stimulating
print('No PubMed results for hypothesis h-var-799795f6af')
No PubMed results for hypothesis h-var-799795f6af
Hypothesis 13: TREM2-SIRT1 Metabolic Senescence Circuit in Microglial Aging¶
Target genes: TREM2 · Composite score: 0.882
This hypothesis proposes that age-related TREM2 signaling dysfunction in microglia triggers cellular senescence through suppression of the SIRT1-dependent metabolic sensing circuit. Under normal conditions, TREM2 activation maintains microglial energy homeostasis by promoting SIRT1-mediated deacetyl
print('No PubMed results for hypothesis h-var-ddd5c9bcc8')
No PubMed results for hypothesis h-var-ddd5c9bcc8
Hypothesis 14: TREM2-Mediated Astrocyte-Microglia Cross-Talk in Neurodegeneration¶
Target genes: TREM2 · Composite score: 0.88
Background and Rationale
While TREM2's role in microglial function is well-established, emerging evidence suggests that TREM2 signaling critically regulates astrocyte-microglia communication networks that become dysregulated in neurodegeneration. Recent studies have identified that microglial T
print('No PubMed results for hypothesis h-var-7e118a66d8')
No PubMed results for hypothesis h-var-7e118a66d8
Hypothesis 15: TREM2-Mediated Astrocyte-Microglia Cross-Talk in Neurodegeneration¶
Target genes: TREM2 · Composite score: 0.875
Background and Rationale
TREM2 dysfunction in Alzheimer's disease extends beyond direct microglial effects to encompass critical astrocyte-microglia communication networks that become dysregulated during neurodegeneration. While TREM2 is exclusively expressed on microglia, emerging evidence sug
print('No PubMed results for hypothesis h-var-a065d9bdf2')
No PubMed results for hypothesis h-var-a065d9bdf2
Hypothesis 16: TREM2-Mediated Cholesterol Dysregulation in Microglial Senescence¶
Target genes: CYP46A1 · Composite score: 0.869
Molecular Mechanism and Rationale¶
The core mechanism centers on TREM2's role in regulating cholesterol homeostasis through modulation of CYP46A1 expression, where TREM2 deficiency disrupts the normal cholesterol efflux pathway in microglia. Under physiological conditions, TREM2 signaling through
print('No PubMed results for hypothesis h-var-e0e82ff2e2')
No PubMed results for hypothesis h-var-e0e82ff2e2
Hypothesis 17: Early Proteasome Restoration Therapy¶
Target genes: PSMC · Composite score: 0.712
Molecular Mechanism and Rationale¶
The 26S proteasome represents the primary degradation machinery for misfolded and damaged proteins in eukaryotic cells, comprising a 20S catalytic core particle flanked by two 19S regulatory particles. The PSMC (Proteasome 26S Subunit, ATPase) gene family encode
print('No PubMed results for hypothesis h-9588dd18')
No PubMed results for hypothesis h-9588dd18
Hypothesis 18: Ferroptosis Inhibition for α-Synuclein Neuroprotection¶
Target genes: GPX4 · Composite score: 0.705
Molecular Mechanism and Rationale¶
Ferroptosis represents a distinct form of regulated cell death characterized by iron-dependent lipid peroxidation and subsequent membrane damage, fundamentally different from apoptosis, necrosis, or autophagy. The central molecular mechanism revolves around the
print('No PubMed results for hypothesis h-2c776894')
No PubMed results for hypothesis h-2c776894
Hypothesis 19: cGAS-STING Senescence Circuit Disruption¶
Target genes: CGAS, STING1 · Composite score: 0.691
Molecular Mechanism and Rationale¶
The cyclic GMP-AMP synthase (cGAS) and stimulator of interferon genes (STING) pathway represents a fundamental innate immune sensing mechanism that has emerged as a critical driver of age-related neurodegeneration. This cytosolic DNA sensing cascade, originally
print('No PubMed results for hypothesis h-bbe4540f')
No PubMed results for hypothesis h-bbe4540f
Hypothesis 20: White Matter Oligodendrocyte Protection via CXCL10 Inhibition¶
Target genes: CXCL10 · Composite score: 0.675
Molecular Mechanism and Rationale¶
The chemokine CXCL10 (C-X-C motif chemokine ligand 10), also known as interferon-γ-inducible protein 10 (IP-10), represents a critical molecular nexus in the pathogenesis of white matter degeneration during aging and neurodegeneration. CXCL10 is a 10 kDa protein
print('No PubMed results for hypothesis h-724e3929')
No PubMed results for hypothesis h-724e3929
Hypothesis 21: Age-Dependent Complement C4b Upregulation Drives Synaptic Vulnerabilit¶
Target genes: C4B · Composite score: 0.671
Age-Dependent Complement C4b Upregulation Drives Synaptic Vulnerability in Hippocampal CA1 Neurons¶
Background & Rationale¶
Aging is the strongest risk factor for Alzheimer's disease and other neurodegenerative conditions, yet the molecular mechanisms linking normal brain aging to neurodegenera
print('No PubMed results for hypothesis h-2f43b42f')
No PubMed results for hypothesis h-2f43b42f
Hypothesis 22: White Matter Vulnerability Prevention via Oligodendrocyte Protection¶
Target genes: CXCL10 · Composite score: 0.667
Molecular Mechanism and Rationale¶
The white matter vulnerability prevention hypothesis centers on a cascade of inflammatory events that compromise oligodendrocyte viability during aging. In this model, age-related microglial activation leads to increased production of C-X-C motif chemokine ligan
print('No PubMed results for hypothesis h-c5698ce3')
No PubMed results for hypothesis h-c5698ce3
Hypothesis 23: Oligodendrocyte White Matter Vulnerability¶
Target genes: MOG · Composite score: 0.651
Background and Rationale
Oligodendrocytes, the myelinating cells of the central nervous system, play a critical role in maintaining neural connectivity and supporting neuronal function. These cells produce myelin sheaths that wrap around axons, facilitating rapid saltatory conduction and provid
print('No PubMed results for hypothesis h-06cb8e75')
No PubMed results for hypothesis h-06cb8e75
Hypothesis 24: Oligodendrocyte Remyelination Enhancement¶
Target genes: TREM2 · Composite score: 0.644
Oligodendrocyte Remyelination Enhancement¶
Mechanistic Hypothesis Overview¶
This hypothesis proposes a disease-modifying strategy centered on Oligodendrocyte Remyelination Enhancement as a mechanistic intervention point in neurodegeneration. The core claim is that the biological process r
print('No PubMed results for hypothesis h-e003a35e')
No PubMed results for hypothesis h-e003a35e
Hypothesis 25: White Matter Immune Checkpoint Restoration¶
Target genes: CXCL10 · Composite score: 0.644
CXCL10 Antagonism to Prevent CD8+ T Cell-Mediated White Matter Degeneration
Overview¶
White matter integrity is essential for cognitive function, enabling rapid signal propagation between brain regions. In aging and neurodegenerative disease, white matter undergoes progressive degradation charac
print('No PubMed results for hypothesis h-245c3e93')
No PubMed results for hypothesis h-245c3e93
Hypothesis 26: Mitochondrial NAD+ Salvage Enhancement¶
Target genes: STING1 · Composite score: 0.639
STING-NAD+ Circuit Modulation for Neuroprotection
Overview¶
NAD+ (nicotinamide adenine dinucleotide) is a central metabolic cofactor required for energy generation, DNA repair, and cellular signaling in all living cells. In the aging brain, NAD+ levels decline by 30-50%, with particularly severe
print('No PubMed results for hypothesis h-3da804f5')
No PubMed results for hypothesis h-3da804f5
Hypothesis 27: Selective Neuronal Vulnerability Network Targeting¶
Target genes: Cell-type specific vulnerability markers · Composite score: 0.638
Molecular Mechanism and Rationale¶
The selective neuronal vulnerability network targeting hypothesis centers on the differential expression of cell-type specific vulnerability markers that render distinct neuronal populations susceptible to age-related degeneration through metabolic stress and co
print('No PubMed results for hypothesis h-0f2b2111')
No PubMed results for hypothesis h-0f2b2111
Hypothesis 28: Selective Cholinergic Protection via APP Pathway Modulation¶
Target genes: APP · Composite score: 0.629
Selective Cholinergic Protection via APP Pathway Modulation¶
Mechanistic Hypothesis Overview¶
The "Selective Cholinergic Protection via APP Pathway Modulation" hypothesis proposes that the selective vulnerability of basal forebrain cholinergic neurons in Alzheimer's disease arises from their
print('No PubMed results for hypothesis h-0d576989')
No PubMed results for hypothesis h-0d576989
Hypothesis 29: Myelin Sulfatide Restoration¶
Target genes: GAL3ST1 · Composite score: 0.623
Myelin Sulfatide Restoration¶
Mechanistic Hypothesis Overview¶
This hypothesis proposes a disease-modifying strategy centered on Myelin Sulfatide Restoration as a mechanistic intervention point in neurodegeneration. The core claim is that the biological process represented by myelin sulfa
print('No PubMed results for hypothesis h-d9604ebf')
No PubMed results for hypothesis h-d9604ebf
Hypothesis 30: TFEB-PGC1α Mitochondrial-Lysosomal Decoupling¶
Target genes: TFEB · Composite score: 0.622
Background and Rationale
The transcription factor EB (TFEB) serves as the master regulator of the coordinated lysosomal expression and regulation (CLEAR) network, controlling the biogenesis and function of lysosomes and autophagosomes. Simultaneously, peroxisome proliferator-activated receptor
print('No PubMed results for hypothesis h-e5a1c16b')
No PubMed results for hypothesis h-e5a1c16b
Hypothesis 31: Microglial ACE Enhancement for Amyloid Clearance¶
Target genes: ACE · Composite score: 0.622
Background and Rationale
Alzheimer's disease (AD) represents a complex neurodegenerative disorder characterized by progressive cognitive decline, with amyloid-β (Aβ) plaques serving as one of the defining pathological hallmarks. While the amyloid cascade hypothesis has dominated therapeutic dev
print('No PubMed results for hypothesis h-1e28311b')
No PubMed results for hypothesis h-1e28311b
Hypothesis 32: Mitochondrial-Cytokine Axis Modulation¶
Target genes: Mitochondrial respiratory complexes and inflammatory cytokine receptors · Composite score: 0.616
Molecular Mechanism and Rationale¶
Age-related neuroinflammation creates a toxic microenvironment where pro-inflammatory cytokines, particularly TNF-α, IL-1β, and IL-6, directly impair mitochondrial function through multiple convergent pathways. These cytokines activate NF-κB and JNK signaling ca
print('No PubMed results for hypothesis h-7dfdc5d7')
No PubMed results for hypothesis h-7dfdc5d7
Hypothesis 33: Complement-Mediated Synaptic Pruning Dysregulation¶
Target genes: C1QA · Composite score: 0.612
Background and Rationale
Synaptic pruning, the selective elimination of synaptic connections, is a fundamental neurodevelopmental process that continues throughout life to maintain optimal neural circuit function. The complement cascade, traditionally recognized as an innate immune system compo
print('No PubMed results for hypothesis h-a8165b3b')
No PubMed results for hypothesis h-a8165b3b
Hypothesis 34: AP1S1-Mediated Vesicular Transport Restoration¶
Target genes: AP1S1 · Composite score: 0.588
Molecular Mechanism and Rationale¶
The AP1S1 protein serves as the sigma subunit of the AP-1 adaptor complex, which is essential for clathrin-mediated vesicular transport between the trans-Golgi network and endosomes. During aging, transcriptional downregulation of AP1S1 compromises the structura
print('No PubMed results for hypothesis h-4639c944')
No PubMed results for hypothesis h-4639c944
Hypothesis 35: TNFRSF25-Mediated Aging Exosome Pathway Inhibition¶
Target genes: TNFRSF25 · Composite score: 0.587
Molecular Mechanism and Rationale¶
The TNFRSF25-mediated aging exosome pathway represents a novel intercellular communication mechanism whereby brain-derived extracellular vesicles carrying age-associated damage signals activate tumor necrosis factor receptor superfamily member 25 (TNFRSF25) on r
print('No PubMed results for hypothesis h-678435d0')
No PubMed results for hypothesis h-678435d0
Hypothesis 36: Senescence-Tau Decoupling Therapy¶
Target genes: CDKN2A · Composite score: 0.585
CDK2A/p16 Inhibition to Break Tau-Senescence Feedback Loop
Overview¶
Cellular senescence and tau pathology are two hallmarks of Alzheimer's disease that have long been studied independently. Emerging evidence reveals a vicious feedback loop between them: tau pathology induces cellular senescence
print('No PubMed results for hypothesis h-08a79bc5')
No PubMed results for hypothesis h-08a79bc5
Hypothesis 37: NOMO1-Mediated Neuronal Resilience Enhancement¶
Target genes: NOMO1 · Composite score: 0.584
Molecular Mechanism and Rationale¶
NOMO1 (Nodal modulator 1) functions as a critical regulator of endoplasmic reticulum (ER) homeostasis through its interaction with the ER membrane protein complex and calcium handling machinery. The protein contains multiple transmembrane domains that facilitate
print('No PubMed results for hypothesis h-9a721223')
No PubMed results for hypothesis h-9a721223
Hypothesis 38: Profilin-1 Cytoskeletal Checkpoint Enhancement¶
Target genes: PFN1 · Composite score: 0.554
Background and Rationale
Microglia, the resident immune cells of the central nervous system, play critical roles in maintaining brain homeostasis through synaptic pruning, debris clearance, and neuronal support. During aging and neurodegenerative diseases, microglia undergo phenotypic changes c
print('No PubMed results for hypothesis h-cd49366c')
No PubMed results for hypothesis h-cd49366c
Hypothesis 39: CD300f Immune Checkpoint Activation¶
Target genes: CD300F · Composite score: 0.545
CD300f Agonism to Restore Aging Brain Immune Balance
Overview¶
The aging brain undergoes a profound transformation in its immune landscape, shifting from a state of balanced vigilance to one of chronic, maladaptive inflammation. Central to this dysregulation is the loss of inhibitory immune chec
print('No PubMed results for hypothesis h-7857b01b')
No PubMed results for hypothesis h-7857b01b
8. Knowledge graph edges (250 total)¶
edge_data = [{'source': 'cGAS-STING pathway activation', 'relation': 'causes (age-related activ', 'target': 'microglial senescence', 'strength': 0.85}, {'source': 'microglial senescence', 'relation': 'causes (creates a feed-fo', 'target': 'neurodegeneration vulnerabilit', 'strength': 0.85}, {'source': 'microglial activation', 'relation': 'causes (microglial activa', 'target': 'CXCL10 production', 'strength': 0.85}, {'source': 'CD8+ T cell recruitment', 'relation': 'causes (recruited CD8+ T ', 'target': 'white matter degeneration', 'strength': 0.85}, {'source': 'aging', 'relation': 'causes (aging causes earl', 'target': 'oligodendrocyte dysfunction', 'strength': 0.85}, {'source': 'CXCL10 inhibition', 'relation': 'causes (CXCL10 antagonist', 'target': 'white matter preservation', 'strength': 0.85}, {'source': 'CXCL10', 'relation': 'causes (CXCL10 acts as ch', 'target': 'CD8+ T cell recruitment', 'strength': 0.85}, {'source': 'ACE enhancement', 'relation': 'causes (microglial ACE en', 'target': 'spleen tyrosine kinase signali', 'strength': 0.82}, {'source': 'ACE enhancement', 'relation': 'causes (enhanced ACE expr', 'target': 'amyloid-β clearance', 'strength': 0.82}, {'source': 'CD8+ T cell recruitment', 'relation': 'causes (recruited CD8+ T ', 'target': 'oligodendrocyte damage', 'strength': 0.8}, {'source': 'aging-activated microglia', 'relation': 'causes (aging activation ', 'target': 'CXCL10 production', 'strength': 0.8}, {'source': 'h-bbe4540f', 'relation': 'implicated_in', 'target': 'neurodegeneration', 'strength': 0.8}, {'source': '27-hydroxycholesterol', 'relation': 'causes (27-hydroxycholest', 'target': 'oligodendrocyte maturation', 'strength': 0.8}, {'source': 'microglial CXCL10 production', 'relation': 'causes (microglia activat', 'target': 'CD8+ T cell recruitment', 'strength': 0.8}, {'source': 'h-bbe4540f', 'relation': 'targets', 'target': 'CGAS', 'strength': 0.8}, {'source': 'h-bbe4540f', 'relation': 'targets', 'target': 'STING1', 'strength': 0.8}, {'source': 'proteostasis failure', 'relation': 'causes (proteostasis fail', 'target': 'neurodegeneration', 'strength': 0.78}, {'source': 'cytokine secretion', 'relation': 'causes (age-related cytok', 'target': 'mitochondrial metabolism suppr', 'strength': 0.78}, {'source': 'proteasome dysfunction', 'relation': 'causes (early proteasome ', 'target': 'proteostasis failure', 'strength': 0.78}, {'source': 'mitochondrial metabolism suppr', 'relation': 'causes (suppressed mitoch', 'target': 'energy stress vulnerability', 'strength': 0.78}, {'source': 'myelin sulfatide deficiency', 'relation': 'causes (loss of sulfatide', 'target': 'microglial activation', 'strength': 0.75}, {'source': 'h-2c776894', 'relation': 'targets', 'target': 'GPX4', 'strength': 0.75}, {'source': 'h-2c776894', 'relation': 'implicated_in', 'target': 'neurodegeneration', 'strength': 0.75}, {'source': 'h-9588dd18', 'relation': 'implicated_in', 'target': 'neurodegeneration', 'strength': 0.75}, {'source': 'AP1S1 downregulation', 'relation': 'causes (age-related downr', 'target': 'clathrin-mediated vesicular tr', 'strength': 0.75}]
if edge_data:
pd.DataFrame(edge_data).head(25)
else:
print('No KG edge data available')
Caveats¶
This notebook uses real Forge tool calls from live APIs:
- Enrichment is against curated gene-set libraries (Enrichr)
- STRING/Reactome/HPA/MyGene reflect curated knowledge
- PubMed literature is search-relevance ranked, not systematic review
The cached evidence bundle is the minimum viable real-data analysis for this topic.