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: FR-PO018

Capturing Spatial Transcriptomics Around Cysts with Deep Learning and Image Processing Is a Novel Method to Reveal Therapeutic Targets in ADPKD

Session Information

Category: Augmented Intelligence, Digital Health, and Data Science

  • 300 Augmented Intelligence, Digital Health, and Data Science

Authors

  • Uesumi, Yoshifumi, Chugai Seiyaku Kabushiki Kaisha, Chuo-ku, Tokyo, Japan
  • Ichida, Yasuhiro, Chugai Seiyaku Kabushiki Kaisha, Chuo-ku, Tokyo, Japan
  • Nagano, Kohji, Chugai Seiyaku Kabushiki Kaisha, Chuo-ku, Tokyo, Japan
  • Horiba, Naoshi, Chugai Seiyaku Kabushiki Kaisha, Chuo-ku, Tokyo, Japan
Background

There are still many unknowns in ADPKD. Current therapies are not sufficiently effective. To search for new therapeutic targets, we examined the mechanism of PKD progression in pericystic areas using a new method that integrates histology, proteomics, spatial transcriptomics, GO (gene ontology) analysis, analysis of DEG (differentially expressed genes), deep learning, and image processing.

Methods

DBA/2J as a normal control mouse and DBA/2FG-pcy as a PKD model mouse were used for the analysis. Frozen kidneys of mice were subjected to proteomic measurement by LC-MS. Spatial transcriptomic measurement (Visium, 10X Genomics) and H&E staining were performed using formalin-fixed, paraffin-embedded samples. Image regions of tissues and cysts were obtained with a deep learning program. The pericystic tissue area was further clarified with image processing. Barcoded spots of Visium in this area were fixed by superimposing the spot locations over the images of the pericystic areas and genetically classifying them with GO analysis or cluster analysis. Barcoded spots were also obtained in the corresponding area in normal control mice to obtain DEG in pericystic areas. GO terms were organized for new criteria to exclude inflammatory or fibrosis genes which are usually upregulated in PKD tissues and include membrane protein genes.

Results

In a mouse model of PKD, proteomics and spatial transcriptomics suggested that genes and proteins were generally present in a constant ratio. We identified the pericystic tissue areas with methods described above. A group of genes was identified by DEG analysis of Barcoded spots in these areas. A list of histopathology-related genes was created by filtering these genes through pre-designed selection criteria.

Conclusion

We created a new list of genes associated with pericystic regions but unrelated to all-kidney tissue that could be involved in PKD progression. We will conduct more detailed investigations on the genes in this list to identify therapeutic targets.