Functional Transcription Factor Target Networks Illuminate Control of Epithelial Remodelling

Abstract

Cell identity is governed by gene expression, regulated by transcription factor (TF) binding at cis-regulatory modules. Decoding the relationship between TF binding patterns and gene regulation is nontrivial, remaining a fundamental limitation in understanding cell decision-making. We developed the NetNC software to predict functionally active regulation of TF targets; demonstrated on nine datasets for the TFs Snail, Twist, and modENCODE Highly Occupied Target (HOT) regions. Snail and Twist are canonical drivers of epithelial to mesenchymal transition (EMT), a cell programme important in development, tumour progression and fibrosis. Predicted “neutral” (non-functional) TF binding always accounted for the majority (50% to 95%) of candidate target genes from statistically significant peaks and HOT regions had higher functional binding than most of the Snail and Twist datasets examined. Our results illuminated conserved gene networks that control epithelial plasticity in development and disease. We identified new gene functions and network modules including crosstalk with notch signalling and regulation of chromatin organisation, evidencing networks that reshape Waddington’s epigenetic landscape during epithelial remodelling. Expression of orthologous functional TF targets discriminated breast cancer molecular subtypes and predicted novel tumour biology, with implications for precision medicine. Predicted invasion roles were validated using a tractable cell model, supporting our approach.

Induced Torpor as a Countermeasure for Low Dose Radiation Exposure in a Zebrafish Model

“…NetNC [ 45 ] analysis of DE genes for the radiation (28.5-rad vs 28.5-Ctrl) and torpor + radiation (18.5-mel-rad vs 28.5-Ctrl) groups produced two networks for each condition (28.5-rad_NET_UP, 28.5-rad_NET_DOWN, 18.5-mel-rad_NET_UP, 18.5-mel-rad_NET_DOWN). Merging of the networks in Cytoscape produced an integrated network (as seen in Figure 10 ) (cytoscape network available in Supplementary Network S1 ) revealing considerable overlap in the NetNC results for each condition and producing two networks that reveal common response mechanisms.…”

“…A zebrafish functional gene network (ZFGN) was kindly provided by Olga Troyanskaya [ 44 ]. The top 500 upregulated and 500 downregulated genes (FC > ±1.5) for the radiation (28.5-rad) and torpor + radiation (18.5-mel-rad) were analysed by NetNC [ 45 ] in ‘Functional Target Identification’ (FTI) mode using the network formed from all ZFGN edges that had edge weight >0.5. The resulting networks for each condition were merged and subsequently visualised using Cytoscape v3.7.2 [ 46 ] and a gene ontology analysis was performed on distinct clusters using the Cytoscape plugin ‘BiNGO’ [ 47 ].…”

Immune Cell Networks Uncover Candidate Biomarkers of Melanoma Immunotherapy Response

“…B-net has the most unique genes ( n = 1022) and the highest average degree (22.8). We derived focus networks for each immune cell grouping by taking the IMMUNETS genes given in Table 1 as input for the NetNC algorithm and using HumanNet as the reference network [ 26 , 27 ] ( Figure 2 , Supplementary Figures S1–S3 and Data Files S2 and S3 ). As expected, the focus networks have connections between clusters that are annotated with biologically related terms.…”

Immune cell networks uncover candidate biomarkers of cancer immunotherapy response

“…Genes from each of the five correlation networks for the regulated genes were taken as input to the NetNC algorithm if they were found in no more than two IMMUNETS networks; in order to produce five focus networks. NetNC analysis used the FTI setting and HumanNet as the base network [28,29]. The five focus networks output from NetNC-FTI were visualized by Cytoscape and annotated with the BiNGO plugin using a significance threshold of q<0.05 [102,103] (Supplementary Data File S3), all of the expressed genes in IRIS [21] were taken as the background gene list for enrichment analysis.…”

“…NK-net is the largest overall (2748 genes) and has the greatest overlap with the other networks, for example sharing 751 and 535 genes with Dend-net and B-net, respectively. In order to refine the IMMUNETS correlation networks, we derived focus networks taking the genes from Table 1 for each immune cell type as input for the NetNC algorithm, and using HumanNet as the reference network [28,29] (Figure 2, Supplementary Figures S1, S2, S3, Supplementary Data Files S2, S3). As expected, the focus networks have connections between clusters that are annotated with biologically related terms.…”