Back

Hierarchical Machine Learning Uncovers Topological Signatures of Autophagy Regulation by Oral Bacteria in Oral Squamous Cell Carcinoma

Latifi-Navid, H.; Akhavan, M.; Jalali, P.; Barzegar Behrooz, A.; Vakili, S.; Vitorino, R.; Beheshti, I.; Menon, A.; Mutalik, V. S.; Schroth, R. J.; Chelikani, P.; Ghavami, S.

2025-12-29 cancer biology
10.64898/2025.12.29.696881 bioRxiv
Show abstract

Oral squamous cell carcinoma (OSCC) progression has been increasingly linked to dysbiosis of the oral microbiome. We hypothesized that pathogenic versus commensal bacteria differentially rewire host autophagy networks to either promote or inhibit OSCC progression. To test this, we constructed host-bacterium autophagy interactomes from KEGG, STRING, and curated databases, identifying key network hubs (e.g., MAPK1, STAT3) via graph-theoretic metrics. We then applied a hierarchical unsupervised machine learning pipeline, combining two-stage principal component analysis with permutation testing and linear discriminant analysis (LDA), to interrogate differences in network topology. This multi-layer approach revealed a clear separation between pro-cancer (pathogenic) and anti-cancer (commensal) bacterial network signatures, with Fusobacterium nucleatum and Streptococcus mitis emerging as dominant global outliers. Pathogenic taxa activated inflammatory-metabolic autophagy signatures (e.g., NFKB1, MYC, ACACA), whereas commensals stabilized kinase-homeostasis signaling (EGFR, PTEN, HSP90AA1). Permutation testing confirmed that these network differences were highly significant and non-random (p < 0.001). We also derived a Dysbiosis Index that robustly distinguished the pro- versus anti-cancer bacterial cohorts with high predictive power. Collectively, our findings highlight oral microbiota-autophagy network topologies as potential biomarkers of OSCC dysbiosis and as novel therapeutic targets. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=189 SRC="FIGDIR/small/696881v1_ufig1.gif" ALT="Figure 1"> View larger version (53K): org.highwire.dtl.DTLVardef@13eb27forg.highwire.dtl.DTLVardef@138bccborg.highwire.dtl.DTLVardef@1f2e651org.highwire.dtl.DTLVardef@1eee140_HPS_FORMAT_FIGEXP M_FIG C_FIG Lay summaryHealthy mouth bacteria help cells stay balanced and protected. When harmful bacteria take over, they disrupt cell recycling (autophagy), increase inflammation, and causing cells to become more aggressive, which can promote oral cancer development.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

1
npj Biofilms and Microbiomes
58 papers in training set
Top 0.1%
31.4%
2
Cell Communication and Signaling
51 papers in training set
Top 0.1%
6.3%
3
Genome Medicine
183 papers in training set
Top 0.7%
4.9%
4
Molecular Oncology
55 papers in training set
Top 0.1%
4.9%
5
Nature Communications
5641 papers in training set
Top 32%
4.1%
50% of probability mass above
6
Cell Reports
1498 papers in training set
Top 12%
3.3%
7
Gut Microbes
78 papers in training set
Top 0.5%
3.3%
8
PLOS Computational Biology
1863 papers in training set
Top 10%
3.3%
9
mBio
833 papers in training set
Top 6%
2.8%
10
eLife
5828 papers in training set
Top 38%
2.7%
11
npj Systems Biology and Applications
125 papers in training set
Top 0.8%
2.1%
12
Science Advances
1243 papers in training set
Top 21%
1.5%
13
PLOS Pathogens
820 papers in training set
Top 7%
1.4%
14
mSphere
302 papers in training set
Top 5%
1.3%
15
Scientific Reports
3612 papers in training set
Top 61%
1.3%
16
PLOS ONE
5266 papers in training set
Top 52%
1.3%
17
Advanced Science
286 papers in training set
Top 7%
1.1%
18
Proceedings of the National Academy of Sciences
2444 papers in training set
Top 34%
1.1%
19
Communications Biology
993 papers in training set
Top 25%
1.0%
20
PNAS Nexus
159 papers in training set
Top 3%
0.9%
21
Genome Research
468 papers in training set
Top 6%
0.9%
22
Molecular & Cellular Proteomics
158 papers in training set
Top 1%
0.9%
23
Microbiome
154 papers in training set
Top 2%
0.9%
24
Frontiers in Microbiology
427 papers in training set
Top 9%
0.6%
25
Cell
431 papers in training set
Top 11%
0.6%
26
International Journal of Cancer
49 papers in training set
Top 1%
0.6%
27
JCI Insight
277 papers in training set
Top 8%
0.6%