A team of researchers has created a pipeline for analysis that seeks out previously unrecognized genes related to specific diseases using five databases that contain gene-disease connections. They applied their pipeline to investigate oxidative stress and its connection to Parkinson’s disease as a practical example.
A team of researchers from Hiroshima University has developed an analytical pipeline to uncover previously overlooked genes associated with specific diseases using five databases that hold gene-disease relationships. Their pipeline was utilized to investigate oxidative stress and its connection to Parkinson’s disease in a case study.
The findings from their research were published in the journal npj Parkinson’s Disease on August 17, 2024.
Scientists can obtain data on genes associated with human diseases from several databases, including the Open Targets Platform, DisGeNET, miRTex, RNADisease, and PubChem. However, these databases sometimes lack certain entries due to errors in curation, biases, and failures in text mining. Additionally, due to the vast amount of research on human diseases, there are challenges in maintaining comprehensive data. Researchers acknowledge that gaps in essential data hinder knowledge-sharing and need to be addressed. Therefore, the research team developed an analytical pipeline aimed at discovering these missing entries for unrecognized genes in the human disease-related gene databases.
“We developed a method to identify novel candidate genes linked to oxidative stress in relation to diseases by using public databases,” stated Hidemasa Bono, a professor at the Laboratory of Genome Informatics and the Laboratory of Bio-DX at Hiroshima University. Their approach consists of three main steps. Firstly, they pinpointed genes responsive to both Parkinson’s disease and oxidative stress by analyzing gene expression data along with transcriptome-wide association study results. Secondly, they accessed multiple public databases that document associations between genes and diseases, focusing on genes that had not previously been linked to Parkinson’s disease. Finally, they refined the selection to emphasize those genes considered more significant by revisiting the gene expression and transcriptome-wide association study data.
“In this research, using oxidative stress and its connection to Parkinson’s disease as a case study, we aimed to find methods to identify genes that have been under-researched yet hold potential research value,” Bono explained. Their analysis brought to light two overlooked genes: nuclear protein 1 (NUPR1) and ubiquitin-like with PHD and ring finger domains 2 (UHRF2). This pipeline stands to assist researchers in uncovering underrepresented genes associated with diseases, thereby facilitating easier access to potentially relevant functional genes.
While searching for genes involved with Parkinson’s disease related to oxidative stress, the team filtered down from 62,226 genes to 168 that showed changes in gene expression in both Parkinson’s disease and oxidative stress scenarios. They then categorized these 168 genes into those linked to Parkinson’s disease and those unassociated based on existing evidence.
From the genes not linked to Parkinson’s disease, they further distilled the list to identify 12 unexploited candidate genes. After a careful examination of these 12 genes, they found that NUPR1 and UHRF2 were unexploited, meaning they were not included in current gene-disease association databases.
The researchers indicated that there are several reasons for the missing entries in gene databases. One possibility is that these databases haven’t been updated recently. Another reason could be failures in text mining extraction. Lastly, some databases depend on expert manual input for new information, which may introduce human errors or biases. Nevertheless, the newly created analytical pipeline to identify overlooked genes helps address these database limitations.
The 12 genes identified in this research respond to both Parkinson’s disease and oxidative stress, yet they have not been thoroughly investigated in Parkinson’s disease-related studies. Understanding these genes could enhance insight into the mechanisms underpinning oxidative stress and Parkinson’s disease and could also be pivotal in developing new treatment strategies. “Moreover, the approach we introduced can also be utilized to narrow down candidate genes related to oxidative stress in other diseases. This may contribute to the advancement of research on oxidative stress and various illnesses,” noted Bono.
The research group includes Takayuki Suzuki and Hidemasa Bono from Hiroshima University and the Research Organization of Information and Systems (RIOS).
The study received backing from the Center of Innovation for Bio-Digital Transformation, the open innovation platform for industry-academia collaboration (COI-NEXT), the Japan Science and Technology Agency, and the ROIS-DS-JOINT.
