# Computational pathology for proteinuric glomerulopathies

> **NIH NIH R01** · UNIVERSITY OF PENNSYLVANIA · 2020 · $203,183

## Abstract

ABSTRACT
The presently employed morphology-based classification system used to distinguish unique glomerular
diseases is inadequate. This classification system is not derived from an understanding of the
molecular basis of these diseases, nor does it well predict observed heterogeneous disease natural
history or response to therapy. This project will apply quantitative pathology methods to assess the
renal parenchyma as one approach to overcoming this critical barrier to glomerular disease clinical
research and patient care. Our goal is to create a well characterized data set that extracts an array of
multiple discrete morphological elements from the structural complexity inherent in glomerular
diseases. We will exploit this detailed characterization—combined with renal morphometrics and
transcriptional and genomic profiling—to explore the hypothesis that inherent in this structural
complexity is extractable information that is predictive of underlying disease biology or clinical outcome.
The proposed study will use existing detailed clinical data and biomaterials obtained by NEPTUNE, a
prospective longitudinal cohort study of proteinuric patients with MCD, FSGS, or MN enrolled at the
time of clinically indicated renal biopsy. As part of the NEPTUNE effort, we created the NEPTUNE
Digital Pathology Protocol (NDPP), an innovative program that (1) compiled a centralized internet-
based digital pathology repository of whole slide digital images, and electron and immunofluorescence
micrographs from 500 NEPTUNE subjects, that (2) allows standardization by application of computer-
based imaging analytical software for multi-level annotation (enumeration) of all glomeruli across all
biopsy levels available, and that (3) enables detailed assessment of renal biopsies using a novel
quantitative scoring system. The proposed project will be carried out in three specific aims: We will (1)
compile a standardized comprehensive morphologic and morphometric profile of 500 NEPTUNE cohort
renal biopsies to test whether these profiles contain reproducible descriptors or groups of descriptors
that can be combined with morphometric measures to reliably assess glomerular disease pathology; (2)
using a variety of statistical approaches, we will identify quantitative structural parameters that are
predictive of outcomes to establish a clinically relevant categorization of proteinuric diseases; and we
will (3) identify transcriptomic signatures, molecular pathways, genetic variants, and expression
quantitative trait loci (eQTL) that associate with specific morphologic and morphometric profiles.

## Key facts

- **NIH application ID:** 9966988
- **Project number:** 5R01DK118431-03
- **Recipient organization:** UNIVERSITY OF PENNSYLVANIA
- **Principal Investigator:** Laura Mariachiara Barisoni
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $203,183
- **Award type:** 5
- **Project period:** 2018-08-03 → 2021-08-29

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9966988

## Citation

> US National Institutes of Health, RePORTER application 9966988, Computational pathology for proteinuric glomerulopathies (5R01DK118431-03). Retrieved via AI Analytics 2026-05-26 from https://api.ai-analytics.org/grant/nih/9966988. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*