The Omics‐Driven Machine Learning Path to Cost‐Effective Precision Medicine in Chronic Kidney Disease

PROTEOMICS, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 10, 2025

ABSTRACT Chronic kidney disease (CKD) poses a significant and growing global health challenge, making early detection slowing progression essential for improving patient outcomes. Traditional diagnostic methods such as glomerular filtration rate proteinuria are insufficient to capture the complexity of CKD. In contrast, omics technologies have shed light on molecular mechanisms CKD, helping identify biomarkers assessment management. Artificial intelligence (AI) machine learning (ML) could transform CKD care, enabling biomarker discovery diagnosis risk prediction, personalized treatment. By integrating multi‐omics datasets, AI can provide real‐time, patient‐specific insights, improve decision support, optimize cost efficiency by avoidance unnecessary treatments. Multidisciplinary collaborations sophisticated ML advance therapeutic strategies in This review presents comprehensive overview pipeline translating data into treatment, covering recent advances research, role critical need clinical validation AI‐driven discoveries ensure their efficacy, relevance, cost‐effectiveness care.

Language: Английский

---

Kidney tea [Orthosiphon aristatus (Blume) Miq.] improves diabetic nephropathy via regulating gut microbiota and ferroptosis

Frontiers in Pharmacology, Journal Year: 2024, Volume and Issue: 15

Published: June 19, 2024

Introduction Diabetic nephropathy (DN) is the leading cause of end-stage renal disease. Due to its complex pathogenesis, new therapeutic agents are urgently needed. Orthosiphon aristatus (Blume) Miq., commonly known as kidney tea, widely used in DN treatment China. However, mechanisms have not been fully elucidated. Methods We db/db mice model and evaluated efficacy tea by measuring fasting blood glucose (FBG), serum inflammatory cytokines, injury indicators histopathological changes. Furthermore, 16S rDNA gene sequencing, untargeted metabolomics, electron microscope, ELISA, qRT-PCR, Western blotting were performed explore which exerted effects. Results Twelve polyphenols identified from extract ameliorated FBG, inflammation mice. Moreover, reshaped gut microbiota, reduced abundance Muribaculaceae , Lachnoclostridium Prevotellaceae_UCG-001 Corynebacterium Akkermansia enriched Alloprevotella Blautia Lachnospiraceae_NK4A136_group . Kidney altered levels metabolites pathways such ferroptosis, arginine biosynthesis mTOR signaling pathway. Importantly, improved mitochondrial damage, increased SOD activity, decreased MDA 4-HNE tissues Meanwhile, this functional upregulated GPX4 FTH1 expression downregulated ACSL4 NCOA4 expression, indicating that it could inhibit ferroptosis kidneys. Conclusion Our findings imply can attenuate development modulating microbiota presents a novel scientific rationale for clinical application tea.

Language: Английский

---

Metabolomics in diabetic nephropathy: Unveiling novel biomarkers for diagnosis (Review)

Molecular Medicine Reports, Journal Year: 2024, Volume and Issue: 30(3)

Published: July 3, 2024

Diabetic nephropathy (DN) also known as diabetic kidney disease, is a major microvascular complication of diabetes and leading cause end‑stage renal disease (ESRD), which affects the morbidity mortality patients with diabetes. Despite advancements in care, current diagnostic methods, such determination albuminuria estimated glomerular filtration rate, are limited sensitivity specificity, often only identifying damage after considerable morphological changes. The present review discusses potential metabolomics an approach for early detection management DN. Metabolomics study metabolites, small molecules produced by cellular processes, may provide more sensitive specific tool compared traditional methods. For purposes this review, systematic search was conducted on PubMed Google Scholar recent human studies published between 2011 2023 that used diagnosis has demonstrated metabolic biomarkers to ability detect broad spectrum metabolites high specificity allow earlier better DN, potentially reducing progression ESRD. Furthermore, pathway analysis assesses pathophysiological mechanisms underlying On whole, By providing in‑depth understanding alterations associated could significantly improve detection, enable timely interventions reduce healthcare burdens condition.

Language: Английский

---

Detection of Fast Decliner of Diabetic Kidney Disease Using Chiral Amino Acid Profiling: A Pilot Study

Chemistry & Biodiversity, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 30, 2025

Biomarkers for the prediction of diabetic kidney disease are still unsatisfactory. Although D-amino acids have been shown to reflect conditions, their efficacy in treating (DKD) has not demonstrated. This study explored potential role as progression markers DKD, an aspect addressed previously. We performed comprehensive acid measurements and collected longitudinal estimated glomerular filtration rate (eGFR) data 135 patients. defined fast decliners (FDs) patients exhibiting >10% decline from baseline eGFR per year compared levels FDs non-FDs. Then, we verified that could predict independent creatinine levels. In with disease, D-serine, D-alanine, D-proline were only detected blood, while 15 urine. Using supervised orthogonal partial least squares analysis, blood D-serine urine identified features characterizing disease. Baseline ratios did differ between FD non-FD groups; however, short-term changes differed. emphasized significance a prognostic marker previous research.

Language: Английский

---

Anti-HBV activity of (R)-gentiandiol, a metabolite of Swertiamarin, in transgenic mice: Insights from non-targeted serum metabolomics

Bioorganic & Medicinal Chemistry, Journal Year: 2025, Volume and Issue: 121, P. 118128 - 118128

Published: Feb. 24, 2025

Language: Английский

---

Predicting coronavirus disease 2019 severity using explainable artificial intelligence techniques

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 19, 2025

Predictive models for determining coronavirus disease 2019 (COVID-19) severity have been established; however, the complexity of interactions among factors limits use conventional statistical methods. This study aimed to establish a simple and accurate predictive model COVID-19 using an explainable machine learning approach. A total 3,301 patients ≥ 18 years diagnosed with between February 2020 October 2022 were included. The discovery cohort comprised whose onset fell before 1, (N = 1,023), validation remaining 2,278). Pointwise linear logistic regression used extract 41 features. Reinforcement was generate high accuracy. primary evaluation area under receiver operating characteristic curve (AUC). achieved AUC 0.905 four features: serum albumin levels, lactate dehydrogenase age, neutrophil count. highest value 0.906 (sensitivity, 0.842; specificity, 0.811) in 0.861 0.804; 0.675) cohort. Simple well-structured established, which may aid patient management selection therapeutic interventions.

Language: Английский

---

Pathogenic variants prevalence patients with diabetic kidney disease in Japan: A descriptive study

Journal of Diabetes Investigation, Journal Year: 2025, Volume and Issue: unknown

Published: April 8, 2025

ABSTRACT Aims/Introduction The impact of rare pathogenic variants on diabetic kidney disease (DKD) has not been investigated in detail. Previous studies have detected 22% Caucasian patients with DKD; however, this proportion may vary depending ethnicity and updates to the database. Therefore, we performed a whole‐genome analysis DKD type 2 diabetes mellitus Japan, utilizing recent database investigate prevalence kidney‐related describe characteristics these patients. Materials methods Whole‐genome sequencing was performed, were analyzed following GATK Best Practices. We extracted data 790 genes associated Mendelian genitourinary diseases. Pathogenic defined based American College Medical Genetics criteria, including both heterozygous homozygous classified as or likely pathogenic. Results Among 79 participants, identified 27 (34.1%), higher than previously reported. No detected. roughly divided into 23.7% related glomerulopathy, 36.8% tubulointerstitial disease, 10.5% cystic disease/ciliopathy, 28.9% others. Diagnostic found 10 (12.7%) seven ( ABCC6, ALPL, ASXL1 , BMPR2 GCM2, PAX2 WT1 ), all autosomal dominant congenital disease. Conclusions This study considerable number Japan who carried variants. These findings suggest potential ethnic differences highlight variant detection.

Language: Английский

---

Intervention treatment reducing cellular senescence inhibits tubulointerstitial fibrosis in diabetic mice following acute kidney injury

Clinical Science, Journal Year: 2024, Volume and Issue: 138(5), P. 309 - 326

Published: Feb. 23, 2024

Senescence of kidney tubules leads to tubulointerstitial fibrosis (TIF). Proximal tubular epithelial cells undergo stress-induced senescence during diabetes and episodes acute injury (AKI), combining these injuries promotes the progression diabetic disease (DKD). Since TIF is crucial DKD, we examined therapeutic potential targeting with a senolytic drug (HSP90 inhibitor) and/or senostatic (ASK1 in model which AKI superimposed on diabetes. After 8 weeks streptozotocin-induced diabetes, mice underwent bilateral clamping renal pedicles induce mild AKI, followed by 28 days reperfusion. Groups (n=10-12) received either vehicle, HSP90 inhibitor (alvespimycin), ASK1 (GS-444217), or both treatments. Vehicle-treated displayed at day 3 extensive cell 10, remained unresolved 28. Markers (Cdkn1a Cdkn2a), inflammation (Cd68, Tnf, Ccl2), (Col1a1, Col4a3, α-Sma/Acta2, Tgfb1) were elevated 28, coinciding function impairment. Treatment alvespimycin alone reduced levels Col1a1, Acta2, Tgfb1, Cd68; however, further treatment GS-444217 also Ccl2, Senolytic therapy can inhibit but its effectiveness be improved follow-up inhibitor, has important implications for treating progressive DKD.

Language: Английский

---

Metabolomic profiles in serum and urine uncover novel biomarkers in children with nephrotic syndrome

European Journal of Clinical Investigation, Journal Year: 2023, Volume and Issue: 53(7)

Published: March 1, 2023

Nephrotic syndrome is common in children and adults worldwide, steroid-sensitive nephrotic (SSNS) accounts for 80%. Aberrant metabolism involvement early SSNS sparsely studied, its pathogenesis remains unclear. Therefore, the goal of this study was to investigate changes initiated patients-related metabolites through serum urine metabolomics discover novel potential metabolic pathways.Serum samples (27 56 controls) (17 24 were collected. Meanwhile, non-targeted analyses performed by ultra-high-performance liquid chromatography-quadrupole time flight-mass spectrometry (UHPLC-QTOF-MS) determine SSNS. We applied causal inference model, DoWhy assess effects several selected metabolites. An ultraperformance chromatography-tandem mass (UPLC-MS/MS) used validate hits (D-mannitol, dulcitol, D-sorbitol, XMP, NADPH, NAD, bilirubin, α-KG-like) 41 43 controls. In addition, pathways explored.Compared urine, analysis more clearly discriminated at 194 differential five obtained group. Eight identified establishing diagnostic model SSNS, four variables had a positive effect. After validation targeted MS, except others have similar trends like untargeted analysis.With further quantitative analysis, we found seven may be new biomarkers risk prediction diagnosis

Language: Английский

---

Metabolomic Investigation of Blood and Urinary Amino Acids and Derivatives in Patients with Type 2 Diabetes Mellitus and Early Diabetic Kidney Disease

Biomedicines, Journal Year: 2023, Volume and Issue: 11(6), P. 1527 - 1527

Published: May 25, 2023

Diabetic kidney disease (DKD) is the leading cause of end-stage renal disease; however, few biomarkers its early identification are available. The aim study was to assess new in stages DKD type 2 diabetes mellitus (DM) patients. This cross-sectional pilot performed an integrated metabolomic profiling blood and urine 90 patients with DM, classified into three subgroups according albuminuria stage from P1 P3 (30 normo-, 30 micro-, macroalbuminuric) 20 healthy controls using high-performance liquid chromatography mass spectrometry (UPLC-QTOF-ESI* MS). From a large cohort separated identified molecules, 33 39 amino acids derivatives serum urine, respectively, were selected for statistical analysis Metaboanalyst 5.0. online software. multivariate univariate algorithms confirmed relevance some as that responsible discrimination between Serum molecules such tiglylglycine, methoxytryptophan, serotonin sulfate, 5-hydroxy lysine, taurine, kynurenic acid, tyrosine found be more significant group C P1-P2-P3. In o-phosphothreonine, aspartic uric among most relevant metabolites group, well these potential may indicate their involvement 2DM progression, reflecting injury at specific sites along nephron, even DKD.

Language: Английский

---