ASN's Mission

To create a world without kidney diseases, the ASN Alliance for Kidney Health elevates care by educating and informing, driving breakthroughs and innovation, and advocating for policies that create transformative changes in kidney medicine throughout the world.

learn more

Contact ASN

1401 H St, NW, Ste 900, Washington, DC 20005

email@asn-online.org

202-640-4660

The Latest on X

Kidney Week

Abstract: SA-OR23

Machine Learning-Based Multi-Omics Analyses Predict Disease Severity and Identify Molecular Mechanisms of COVID-Associated Kidney Injury

Session Information

Category: Coronavirus (COVID-19)

  • 000 Coronavirus (COVID-19)

Authors

  • Anandakrishnan, Nanditha, Icahn School of Medicine at Mount Sinai, New York, New York, United States
  • Yi, Zhengzi, Icahn School of Medicine at Mount Sinai, New York, New York, United States
  • Liu, Tong, Rutgers The State University of New Jersey, New Brunswick, New Jersey, United States
  • Sun, Zeguo, Icahn School of Medicine at Mount Sinai, New York, New York, United States
  • Haydak, Jonathan C., Icahn School of Medicine at Mount Sinai, New York, New York, United States
  • Kretzler, Matthias, University of Michigan, Ann Arbor, Michigan, United States
  • Li, Hong, Rutgers The State University of New Jersey, New Brunswick, New Jersey, United States
  • Nadkarni, Girish N., Icahn School of Medicine at Mount Sinai, New York, New York, United States
  • Chan, Lili, Icahn School of Medicine at Mount Sinai, New York, New York, United States
  • Zhang, Weijia, Icahn School of Medicine at Mount Sinai, New York, New York, United States
  • Azeloglu, Evren U., Icahn School of Medicine at Mount Sinai, New York, New York, United States
Background

Studies show a higher prevalence of acute kidney injury (AKI) in COVID-19 patients, however mechanisms leading to severe kidney outcomes are unclear. Identifying these mechanisms may help develop therapies for the management of COVID-associated nephropathy (COVAN). In this multicenter study, we combine urinary proteomics and machine learning (ML) to predict severe kidney outcomes in COVID-19 patients and further combine multiomic datasets to identify gene networks driving disease (Fig 1A).

Methods

Quantitative LC-MS/MS analysis of urine samples from hospitalized participants was performed. For ML algorithm construction, samples were stratified into severe and mild outcomes and randomized into discovery and validation set at a 2:1 ratio. Limma test was used to identify differentially abundant proteins (DAPs) and the Boruta feature selection method was used to select features to construct the ML algorithm.

Results

Limma test on the discovery set identified DAPs in severe vs mild outcome cohort (Fig 1B). The top features identified using Boruta feature selection method used for random forest model construction demonstrated good predictive power of greater than 76% accuracy for both discovery and validation set (Fig 1C, D). Enrichment analysis showed significant upregulation of exocytosis and immune related processes and downregulation of cell adhesion and extracellular matrix organization related processes in severe COVID-19 (Fig 1E, F). The top features showed expression in multiple nephron segments based on the public KPMP data (Fig 1G).

Conclusion

The novel biomarkers identified here can be used for assessment of kidney function in patients with COVAN.

Funding

  • NIDDK Support