↗
Glycolysis activation
The study linked increased glycolytic activity to DNA hypomethylation, connecting an epigenetic change to a metabolic phenotype.
Multi-omics, interpreted biologically
Follow the signal from DNA to RNA to protein.
BioSircle imagines a research platform built around the logic of Signature Regulatory Clustering—organizing dysregulated genes by the biological layer where change first appears.
3omics layers
1biological flow
VAEintegrated ranking
Regulatory flowresearch model
Layer 01DNA methylationEpigenetic control
↓Layer 02RNA expressionTranscriptional response
↓Layer 03Protein abundanceTranslational outcome
DNA methylation×RNA-seq×Proteomics→Regulatory clusters
What the study revealed
Different biological layers can drive different parts of the cancer phenotype.
Applied to clear cell renal cell carcinoma, the method surfaced regulatory patterns that single-layer analyses can miss.
⌁
Mitochondrial suppression
Mitochondrial enzymes and respiratory-chain complexes were found to be suppressed at the translational layer.
◎
Loss of cell identity
Stage-dependent downregulation of proximal renal tubule genes suggested progressive loss of cellular identity in cancer cells.
Why integration matters
Move beyond lists of altered genes.
SiRCle groups genes by the first layer where dysregulation occurs, then combines those clusters with a variational autoencoder to expose biologically meaningful features and compare patient subpopulations.
See the model step by step →Phenotype
DNARNAProteinVAEPrimary source
Built from peer-reviewed, open-access research.
Genome Medicine · 2024 · 16:144
SiRCle (Signature Regulatory Clustering) model integration reveals mechanisms of phenotype regulation in renal cancer
Ariane Mora, Christina Schmidt, Brad Balderson, Christian Frezza & Mikael Bodén
Open full article ↗