Integrating node embeddings and biological annotations for genes to predict disease-gene associations

Background

Predicting disease causative genes (or simply, disease genes) has played critical roles in understanding the genetic basis of human diseases and further providing disease treatment guidelines. While various computational methods have been proposed for disease gene prediction, with the recent increasing availability of biological information for genes, it is highly motivated to leverage these valuable data sources and extract useful information for accurately predicting disease genes.

Results

We present an integrative framework called N2VKO to predict disease genes. Firstly, we learn the node embeddings from protein-protein interaction (PPI) network for genes by adapting the well-known representation learning method node2vec. Secondly, we combine the learned node embeddings with various biological annotations as rich feature representation for genes, and subsequently build binary classification models for disease gene prediction. Finally, as the data for disease gene prediction is usually imbalanced (i.e. the number of the causative genes for a specific disease is much less than that of its non-causative genes), we further address this serious data imbalance issue by applying oversampling techniques for imbalance data correction to improve the prediction performance. Comprehensive experiments demonstrate that our proposed N2VKO significantly outperforms four state-of-the-art methods for disease gene prediction across seven diseases.

Conclusions

In this study, we show that node embeddings learned from PPI networks work well for disease gene prediction, while integrating node embeddings with other biological annotations further improves the performance of classification models. Moreover, oversampling techniques for imbalance correction further enhances the prediction performance. In addition, the literature search of predicted disease genes also shows the effectiveness of our proposed N2VKO framework for disease gene prediction.

     

The metal-coordinated Casiopeína IIIEa induces the petite-like phenotype in <Emphasis Type="Italic">Saccharomyces cerevisiae</Emphasis>

The Casiopeínas^® are mixed chelate copper (II) complexes and promising antineoplastics agents against cancer cells and tumors in vitro and in vivo. However, the action mode of these compounds is poorly characterized. In this work the effect of the antineoplastic Casiopeína IIIEa on the metabolism and ultrastructure of the yeast Saccharomyces cerevisiae was investigated. Exposure of cells growing in rich or in low-iron medium to 5 μM of the compound decreased duplication time and reduced oxygen consumption. Those cells formed smaller colonies when growing in a non-fermentable carbon source and low-iron medium, and under the light microscope, multiple folds were observed along the plasma membrane accompanied with a reduction in the diameter of the yeast. These observations were confirmed under the electron microscope, which also revealed a slight reduction of the mitochondrial size. A correlation was found with smaller colonies exhibiting lower rates of oxygen consumption, and yeast labelled with fluorescent MitoTracker^TM consistently exhibited reduced mitochondrial activity. It appears that Casiopeína IIIEa gives rise to smaller yeast and petite-like colonies by reducing the mitochondrial respiratory activity without significantly affecting the mitochondrial structure.     

Resveratrol ameliorates oxidative DNA damage and protects against acrylamide-induced oxidative stress in rats

Acrylamide (ACR), used in many fields from industrial manufacturing to laboratory personnel work is also formed during the heating process through interactions of amino acids. Therefore ACR poses a significant risk to human health. This study aimed to elucidate whether resveratrol (RVT) treatment could modulate ACR-induced oxidative DNA damage and oxidative changes in rat brain, lung, liver, kidney and testes tissues. Rats were divided into four groups as control (C); RVT (30 mg/kg i.p. dissolved in 0.9% NaCl), ACR (40 mg/kg i.p.) and RVT + ACR groups. After 10 days rats were decapitated and tissues were excised. 8-hydroxydeoxyguanosine (8-OHdG) is a biomarker of oxidative DNA damage. 8-OHdG content in the extracted DNA solution was determined by enzyme-linked immunosorbent assay method. Malondialdehyde (MDA), glutathione (GSH) levels and myeloperoxidase activity (MPO) were determined in tissues, while oxidant-induced tissue fibrosis was determined by collagen contents. Serum enzyme activities, cytokine levels, leukocyte apoptosis were assayed in plasma. As an indicator of oxidative DNA damage, 8-OHdG levels significantly increased in ACR group and this was reversed significantly by RVT treatment. In ACR group, GSH levels decreased significantly while the MDA levels, MPO activity and collagen content increased in the tissues suggesting oxidative organ damage. In RVT-treated ACR group, oxidant responses reversed significantly. Serum enzyme activities, cytokine levels and leukocyte late apoptosis which increased following ACR administration, decreased with RVT treatment. Therefore supplementing with RVT can be useful in individuals at risk of ACR toxicity.