Inferring gene regulatory networks of ALS from blood transcriptome profiles

• Published in: Heliyon Vol. 10; no. 23; p. e40696
• Main Authors: Pappalardo, Xena G., Jansen, Giorgio, Amaradio, Matteo, Costanza, Jole, Umeton, Renato, Guarino, Francesca, De Pinto, Vito, Oliver, Stephen G., Messina, Angela, Nicosia, Giuseppe
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 15.12.2024; Elsevier
• Subjects: Algorithm for the reconstruction of accurate cellular networks - adaptive partitioning version (ARACNe-AP); algorithms; amyotrophic lateral sclerosis; Amyotrophic lateral sclerosis (ALS); biomarkers; blood; data collection; females; gene expression; Gene expression profile (GEP); Gene regulatory network (GRN); gene regulatory networks; genes; humans; phenotype; prediction; Reverse engineering methods; transcription (genetics); transcriptome; Gene expression profile (GEP); Gene regulatory network (GRN); Amyotrophic lateral sclerosis (ALS); Reverse engineering methods; Algorithm for the reconstruction of accurate cellular networks - adaptive partitioning version (ARACNe-AP)
• ISSN: 2405-8440; 2405-8440
• DOI: 10.1016/j.heliyon.2024.e40696

One of the most robust approaches to the prediction of causal driver genes of complex diseases is to apply reverse engineering methods to infer a gene regulatory network (GRN) from gene expression profiles (GEPs). In this work, we analysed 794 GEPs of 1117 human whole-blood samples from Amyotrophic Lateral Sclerosis (ALS) patients and healthy subjects reported in the GSE112681 dataset. GRNs for ALS and healthy individuals were reconstructed by ARACNe-AP (Algorithm for the Reconstruction of Accurate Cellular Networks - Adaptive Partitioning). In order to examine phenotypic differences in the ALS population surveyed, several datasets were built by arranging GEPs according to sex, spinal or bulbar onset, and survival time. The designed reverse engineering methodology identified a significant number of potential ALS-promoting mechanisms and putative transcriptional biomarkers that were previously unknown. In particular, the characterization of ALS phenotypic networks by pathway enrichment analysis has identified a gender-specific disease signature, namely network activation related to the radiation damage response, reported in the networks of bulbar and female ALS patients. Also, focusing on a smaller interaction network, we selected some hub genes to investigate their inferred pathological and healthy subnetworks. The inferred GRNs revealed the interconnection of the four selected hub genes (TP53, SOD1, ALS2, VDAC3) with p53-mediated pathways, suggesting the potential neurovascular response to ALS neuroinflammation. In addition to being well consistent with literature data, our results provide a novel integrated view of ALS transcriptional regulators, expanding information on the possible mechanisms underlying ALS and also offering important insights for diagnostic purposes and for developing possible therapies for a disease yet incurable. •ARACNe-AP tool was applied to reconstruct the gene regulatory networks (GRNs).•ARACNe inferred GRNs from blood transcriptomic data of ALS and healthy individuals.•GRNs based on gender and ALS disease onset type outlined novel deregulated pathways.•Regulatory patterns of some hub genes (TP53, SOD1, ALS2, VDAC3) were characterized.•ARACNe-GRNs identified ALS phenotypic motifs and transcriptional gene signatures.