Akademik Veri Yönetim Sistemi
Aydın Z., Zararsız G.(Yürütücü)
TÜBİTAK Projesi, 2022 - 2024
With more than 12.5 million people worldwide, autosomal dominant polycystic kidney disease (ADPKD) is the most common hereditary kidney disease (1/1000) and is one of the most common life-threatening (monogenic) diseases. Approximately 50% of the patients with ADPKD ultimately progress to end-stage renal disease, patients inevitably undergo dialysis or require kidney transplantation. In addition, there are many studies on the pathophysiology of this disease, which has a heterogeneous clinical course, even in the same family members. Although genomic, transcriptomics and proteomics are used in the elucidation of chronic kidney disease and to explain its molecular mechanisms, it is emphasized that metabolomic applications are an appropriate experimental tool for biomarker and prognosis especially in chronic kidney research. Genomic-transcriptomic-proteomic methods have not yet been used in clinical practice, although relevant “-omics” sciences have very strong and reliable aspects in understanding the disease pathogenesis. Thus, the metabolomics method has been adopted as the technology of choice that will respond to “–omics” level in order to understand the course of chronic kidney diseases such as ADPKD. In this context, a complete demonstration of the metabolomic fingerprint in the plasma specimens of ADPKD, which is caused by systemic (genetic-based) damage in tubular cells in the kidney, instead of biochemical parameters, may help explain the prognosis of the disease. In addition, the determination of the metabolite profile of ADPKD may help to develop current treatment approaches and increase their effectiveness, and carry out the potential support to the previously performed molecular studies. Within the scope of the project, it is aimed to carry out analyzes with comprehensive molecular data in order to predict the progression of the disease in ADPKD patients and to create a decision support system for the follow-up of the disease based on these findings. It was aimed to create a comprehensive R library that includes machine learning and deep learning algorithms to create the infrastructure of the decision support system and to effectively realize the predictive models for tracking. With this decision support system, it is aimed to obtain personalized follow-up reports regarding the disease and thus to develop a result-oriented system where remote patient monitoring can be performed and evaluated. For this purpose, the a multi-disciplinary working team has been set up to complement each other such as nephrology, basic sciences (biology / biochemistry), biostatistics and computer engineering. It is aimed to obtain a prototype for the project output and to start the pilot application in Erciyes University Medical Faculty Nephrology Clinic.