Generating pregnant patient biological profiles by deconvoluting clinical records with electronic health record foundation models
Authors:
- David SeongDSDavid Seong
- Samson MatarasoSMSamson Mataraso
- Camilo EspinosaCECamilo Espinosa
- Eloise BersonEBEloise Berson
- S Momsen ReinckeSRS Momsen Reincke
- Lei XueLXLei Xue
- Chloe KashiwagiCKChloe Kashiwagi
- Yeasul KimYKYeasul Kim
- Chi-Hung ShuCSChi-Hung Shu
- Philip ChungPCPhilip Chung
- Marc GhanemMGMarc Ghanem
- Feng XieFXFeng Xie
Open Access
- 23 September 2024
- journal article
- research article
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 25 (6)
Abstract
Translational biology posits a strong bi-directional link between clinical phenotypes and a patient’s biological profile. By leveraging this bi-directional link, we can efficiently deconvolute pre-existing clinical information into biological profiles. However, traditional computational tools are limited in their ability to resolve this link because of the relatively small sizes of paired clinical–biological datasets for training and the high dimensionality/sparsity of tabular clinical data. Here, we use state-of-the-art foundation models (FMs) for electronic health record (EHR) data to generate proteomics profiles of pregnant patients, thereby deconvoluting pre-existing clinical information into biological profiles without the cost and effort of running large-scale traditional omics studies. We show that FM-derived representations of a patient’s EHR data coupled with a fully connected neural network prediction head can generate 206 blood protein expression levels. Interestingly, these proteins were enriched for developmental pathways, while proteins not able to be generated from EHR data were enriched for metabolic pathways. Finally, we show a proteomic signature of gestational diabetes that includes proteins with established and novel links to gestational diabetes. These results showcase the power of FM-derived EHR representations in efficiently generating biological states of pregnant patients. This capability can revolutionize disease understanding and therapeutic development, offering a cost-effective, time-efficient, and less invasive alternative to traditional methods of generating proteomics.Keywords
Funding Information
- Chambers–Okamura Prematurity Fund
- Roberts Foundation Research Fund
- Hess Research Fund
- Alfred E. Mann Foundation
- March of Dimes
- Burroughs Wellcome Fund (1019816)
- National Institutes of Health (R35GM138353)
This publication has 51 references indexed in Scilit:
- Progranulin expression in advanced human atherosclerotic plaqueAtherosclerosis, 2009
- Soluble ST2 Blocks Interleukin-33 Signaling in Allergic Airway InflammationJournal of Biological Chemistry, 2007
- IL-33 and ST2 comprise a critical biomechanically induced and cardioprotective signaling systemJournal of Clinical Investigation, 2007
- Serum cystatin C in pregnant women: reference values, reliable and superior diagnostic accuracy.2005
- Thrombospondin-2 Is Essential for Myocardial Matrix IntegrityCirculation Research, 2004
- Pre-eclampsia with fetal growth restriction: placental and serum activin A and inhibin A levels.2002
- Concentration in plasma of macrophage inhibitory cytokine-1 and risk of cardiovascular events in women: a nested case-control studyThe Lancet, 2002
- Changes in inhibins and activin secretion in healthy and pathological pregnanciesMolecular and Cellular Endocrinology, 2001
- Discoidin Domain Receptor 1 Tyrosine Kinase Has an Essential Role in Mammary Gland DevelopmentMolecular and Cellular Biology, 2001
- The value of urine screening for glucose at each prenatal visitObstetrics & Gynecology, 1995