Bryson, K. | bioRxiv Science

bioRxiv2019

The breast cancer oncogene IKKε coordinates mitochondrial function and serine metabolism

The I{kappa}B kinase {varepsilon} (IKK{varepsilon}) is a key molecule at the crossroads of inflammation and cancer. Known for its role as an activator of NF{kappa}B and IRF3 signalling leading to cytokine secretion, the kinase is also a breast cancer oncogene, overexpressed in a variety of tumours. However, to what extent IKK{varepsilon} remodels cellular metabolism is currently unknown. Here we used a combination of metabolomics and phosphoproteomics to show that IKK{varepsilon} orchestrates a complex metabolic reprogramming that affects mitochondrial metabolism and serine biosynthesis. Acting independently of its canonical signalling role, IKK{varepsilon} upregulates the serine biosynthesis pathway (SBP) mainly by limiting glucose and pyruvate derived anaplerosis of the TCA cycle. In turn, this elicits activation of the transcription factor ATF4 and upregulation of the SBP genes. Importantly, pharmacological inhibition of the IKK{varepsilon}-induced metabolic phenotype reduces proliferation of breast cancer cells. Finally, we show that in a set of basal ER negative and highly proliferative human breast cancer tumours, IKK{varepsilon} and PSAT1 expression levels are positively correlated corroborating the link between IKK{varepsilon} and the SBP in the clinical context.

cancer biology↗

bioRxiv2019

Adversarial generation of gene expression data

The problem of reverse engineering gene regulatory networks from high-throughput expression data is one of the biggest challenges in bioinformatics. In order to benchmark network inference algorithms, simulators of well-characterized expression datasets are often required. However, existing simulators have been criticized because they fail to emulate key properties of gene expression data. In this study we address two problems. First, we propose mechanisms to faithfully assess the realism of a synthetic gene expression dataset. Second, we design an adversarial simulator of expression data, gGAN, based on a Generative Adversarial Network. We show that our model outperforms existing simulators by a large margin, achieving realism scores that are up to 17 times higher than those of GeneNetWeaver and SynTReN. More importantly, our results show that gGAN is, to our best knowledge, the first simulator that passes the Turing test for gene expression data proposed by Maier et al. (2013).

bioinformatics↗