NANOG-Mediated Senescence Reversal

Re-analysis of Shahini et al. (2021) showing how NANOG protein can reverse cellular aging in muscle cells.

We used Machine Learning to find hidden patterns in 6,613 genes and built interactive tools + AI chatbots to explore the results.

genes
6,613 genes genes
samples
15 samples samples
clusters
8 clusters clusters
timepoints
4 timepoints timepoints

Experimental Design

Model System

Species Icon
SPECIES
Homo sapiens
CELL TYPE
Human skeletal muscle myoblasts
INTERVENTION
NANOG overexpression
Young Myoblasts

Young Myoblasts

3 samples
Early passage cells (baseline young state)
Y n=3
🧬 Young Baseline
Early passage myoblasts (young skeletal muscle progenitors)
Senescent Myoblasts

Senescent Myoblasts

3 samples
Late passage cells (replicative senescence)
S n=3
ā° Senescent
Late passage myoblasts (replicative senescence)
NANOG Treatment

NANOG Treatment

9 samples
Senescent cells + NANOG (5, 10, 15 days)
SN5 n=3
ā±ļø NANOG 5d
Senescent myoblasts + Nanog overexpression for 5 days
SN10 n=3
ā±ļø NANOG 10d
Senescent myoblasts + Nanog overexpression for 10 days
SN15 n=3
ā±ļø NANOG 15d
Senescent myoblasts + Nanog overexpression for 15 days

Key Findings

6,613 genes analyzed across 8 clusters

Cluster 14 contains NANOG (the intervention gene)

843 significant senescence markers identified

1970 Type B genes (dual senescence marker + NANOG responders)

329 early responders show kinetics within 5 days

Mean reversal increases with duration: 7.4% → 22.0%

Clusters Analyzed

Genes were clustered based on co-expression patterns across young, senescent, and NANOG-treated samples. Cluster 14 is particularly interesting as it contains NANOG itself.

Cluster 1

736 genes • 0.0% unknown
šŸ’Ŗ Strong Response
• 2.2% strong responders (16 genes with |FC|>2.0)
Silhouette: 0.417
Strong responders: 2.2% (16 genes)
Key genes:
F2RL2, EREG, TRHDE

Cluster 4

711 genes • 0.0% unknown
šŸ”„ Youth Reversal
• Youth-enriched pattern (S→Y = -0.35)
• 15-day Nanog response: 0.39
Silhouette: 0.307
Strong responders: 0.7% (5 genes)
Key genes:
GPAT3, L1CAM, LAMA5

Cluster 6

1402 genes • 0.0% unknown
šŸ“Š ConsistentšŸ’Ŗ Strong Response
• Consistent response across conditions (variance=0.068)
• 2.1% strong responders (29 genes with |FC|>2.0)
Silhouette: 0.500
Strong responders: 2.1% (29 genes)
Key genes:
C12orf56, SLAMF9, Metazoa_SRP

Cluster 7

426 genes • 0.0% unknown
āš ļø Senescence-Enriched
• Senescence-enriched pattern (S→Y = +0.35)
• 30% response to 15-day treatment
Silhouette: 0.566
Strong responders: 0.2% (1 genes)
Key genes:
TIMP3, H19, FLNB

Cluster 9

836 genes • 0.0% unknown
šŸŽÆ Timepoint-Specific
• 15-day vs 5-day treatment difference: 0.18 (primary=0.46, tertiary=0.28)
Silhouette: 0.343
Strong responders: 1.3% (11 genes)
Key genes:
PDZRN3, FCRLB, COL28A1

Cluster 11

743 genes • 0.0% unknown
šŸ”„ Youth Reversal
• Youth-enriched pattern (S→Y = -0.42, strongest reversal)
• 15-day Nanog response: 0.35
Silhouette: 0.235
Strong responders: 0.5% (4 genes)
Key genes:
MAP3K9, KLF4, TMEM200A

Cluster 14 ⭐

1000 genes • 0.0% unknown
šŸŽÆ Timepoint-SpecificšŸ’Ŗ Strong Response
• Contains NANOG (the intervention gene)
• 15-day vs 5-day treatment difference: 0.20 (primary=0.59, tertiary=0.39)
Silhouette: 0.342
Strong responders: 3.8% (38 genes)
Key genes:
NANOGP8, NANOG, NANOGP1

Cluster 15

759 genes • 0.0% unknown
šŸ“Š Consistent
• Consistent response across conditions (variance=0.062)
Silhouette: 0.437
Strong responders: 1.8% (14 genes)
Key genes:
GPX3, KIAA1755, SCUBE3

Top Priority Genes

Genes ranked by multi-evidence scoring combining network topology, senescence reversal patterns, cluster stability, and statistical significance. Gene types: Type A (reversing markers - therapeutic successes), Type B (off-target responders), Type C (resistant markers - therapy targets).

Gene Priority Scores

Multi-dimensional scoring from network, reversal, stability, statistical, ml analysis evidence

Show:
RankGene
Score
#1SLC7A14
Type B
0.000
#2PRKG1
Type B
0.000
#3FHDC1
Type B
0.000
#4THRB
Type B
0.000
#5CAMK1
Type B
0.000
#6CTSK
Type B
0.000
#7GALNT18
Type B
0.000
#8PPIL6
Type B
0.000
#9LINC00856
Type B
0.000
#10TSGA10
Type B
0.000
#11RP11-327F22.1
Type A
0.000
#12ZNF662
Type B
0.000
#13RP11-567C2.1
Type B
0.000
#14KRT18
Type B
0.000
#15NUDT7
Type B
0.000
#16CCDC144A
Other
0.000
#17GBP5
Type B
0.000
#18TECTA
Type B
0.000
#19PITPNM3
Type B
0.000
#20EREG
Type B
0.000
Showing top 20 of 100 genes

A Note of Appreciation (From Shahini et al. 2021)

To the entire research team, led by Dr. Stelios T. Andreadis, NIH Grant Owner:

Dr. Stelios T. Andreadis, Aref Shahini, Nika Rajabian, Debanik Choudhury, Shahryar Shahini, Kalyan Vydiam, Thy Nguyen, Joseph Kulczyk, Tyler Santarelli, Izuagie Ikhapoh, Yali Zhang, Jianmin Wang, Song Liu, Aimee Stablewski, Ramkumar Thiyagarajan, Kenneth Seldeen, Dr. Bruce R. Troen, Jennifer Peirick, and Dr. Pedro Lei.

We offer our sincere congratulations to this impressive collaborative team on the publication of your research paper. The successful completion of this valuable work, enabled by the funding and scientific vision of the NIH Grant, speaks volumes about the leadership of Dr. Andreadis and the diligence and expertise of every author.

Furthermore, we deeply appreciate your commitment to open science by publicly sharing your data. This practice significantly enhances the impact and reproducibility of your findings.

Thank you for your excellent and transparent contribution to the scientific community.

Help Us Improve

Your feedback helps us build better tools for aging research