Evidence map›Paper›PMID 40542121›Full record

ArticleNPJ digital medicine2025

Few shot learning for phenotype-driven diagnosis of patients with rare genetic diseases.

Emily Alsentzer, Michelle M Li, Shilpa N Kobren, Ayush Noori, Undiagnosed Diseases Network, Isaac S Kohane, Marinka Zitnik

Abstract read
In one paragraph

Article in NPJ digital medicine, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 25 papers.

0numbers the graph read from it
0cells of the map it votes in
25citing papers in PubMed
–field-weighted citation impact
1 · What the graph read from it

What it found

Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.

The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.

2 · The registry

The trial behind it

Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.

Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.

3 · Its place in the literature

Who cites it

25 citing papers in PubMed.

  1. Review
  2. Article
  3. Article
  4. Article
  5. Article
  6. Review
  7. Article
  8. Genetic Diagnosis and Discovery Enabled by Large Language Models.Advanced science (Weinheim, Baden-Wurttemberg, Germany) · 2026
    Article
  9. AI for scientific discovery is a social problem.Patterns (New York, N.Y.) · 2026
    Review
  10. Article
  11. Article
  12. Article
  13. Review
  14. Article
  15. Article
  16. Review
  17. Article
  18. Article
  19. Article
  20. Review
4 · The record

Corrections and comments

PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.

5 · Who and what money

Authors and funding

7 authors.

Emily Alsentzer *Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA.
Michelle M Li *Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA.
Shilpa N KobrenDepartment of Biomedical Informatics, Harvard Medical School, Boston, MA, USA.
Ayush NooriDepartment of Biomedical Informatics, Harvard Medical School, Boston, MA, USA.
Undiagnosed Diseases Network
Isaac S KohaneDepartment of Biomedical Informatics, Harvard Medical School, Boston, MA, USA.
Marinka ZitnikDepartment of Biomedical Informatics, Harvard Medical School, Boston, MA, USA. marinka@hms.harvard.edu.

Funding

Coordinating Center for the Undiagnosed Diseases NetworkU01HG007530 · NHGRI · HARVARD MEDICAL SCHOOL · PI KOHANE, ISAAC S. · 2014 to 2022
$30.2M
Training Program in Bioinformatics and Integrative GenomicsT32HG002295 · NHGRI · MASSACHUSETTS INSTITUTE OF TECHNOLOGY · PI Peter J Park · 2001 to 2026
$15.8M
Pilot of New Technologies to Increase the Genomic Diagnosis of Undiagnosed Disease Network (UDN) PatientsU01HG007709 · NHGRI · BAYLOR COLLEGE OF MEDICINE · PI BACINO, CARLOS A., LEE, BRENDAN · 2014 to 2022
$14.3M
Vanderbilt Center for Undiagnosed Diseases (VCUD) - BiorepositoryU01HG007674 · NHGRI · VANDERBILT UNIVERSITY MEDICAL CENTER · PI COGAN, JOY D, HAMID, RIZWAN · 2014 to 2022
$13.7M
An integrated and diverse genomic medicine program for undiagnosed diseasesU01HG007672 · NHGRI · DUKE UNIVERSITY · PI SHASHI, VANDANA · 2014 to 2022
$13.4M
Center for Integrated Approaches to Undiagnosed DiseasesU01HG007690 · NHGRI · BRIGHAM AND WOMEN'S HOSPITAL · PI LOSCALZO, JOSEPH · 2014 to 2022
$10.4M
Clinical Sequencing Core Facility for the Undiagnosed Diseases Network (UDN)U01HG007942 · NHGRI · BAYLOR COLLEGE OF MEDICINE · PI ENG, CHRISTINE · 2014 to 2021
$10.2M
UCLA clinical site for the investigation of undiagnosed disordersU01HG007703 · NHGRI · UNIVERSITY OF CALIFORNIA LOS ANGELES · PI MARTINEZ-AGOSTO, JULIAN, NELSON, STANLEY F. · 2014 to 2022
$10.1M
Stanford Center for Undiagnosed DiseasesU01HG007708 · NHGRI · STANFORD UNIVERSITY · PI ASHLEY, EUAN A, BERNSTEIN, JONATHAN ADAM · 2014 to 2018
$8.0M
Zebrafish CoreU54NS093793 · NINDS · BAYLOR COLLEGE OF MEDICINE · PI BELLEN, HUGO J · 2015 to 2022
$7.5M
Pacific Northwest Undiagnosed Diseases Network Clinical SiteU01HG010233 · NHGRI · UNIVERSITY OF WASHINGTON · PI DIPPLE, KATRINA M, JARVIK, GAIL PAIRITZ · 2018 to 2022
$5.4M
Clinical Genome Wide Sequencing Core for the Undiagnosed Disease NetworkU01HG007943 · NHGRI · MEDICAL COLLEGE OF WISCONSIN · PI WORTHEY, ELIZABETH A · 2014 to 2017
$4.1M
NCATS NIH HHS U01 TR001395NCATS NIH HHS U01 TR002471NHGRI NIH HHS T32 HG002295NHGRI NIH HHS U01 HG007530NHGRI NIH HHS U01 HG007672NHGRI NIH HHS U01 HG007674NHGRI NIH HHS U01 HG007690NHGRI NIH HHS U01 HG007703NHGRI NIH HHS U01 HG007708NHGRI NIH HHS U01 HG007709NHGRI NIH HHS U01 HG007942NHGRI NIH HHS U01 HG007943NHGRI NIH HHS U01 HG010215NHGRI NIH HHS U01 HG010217NHGRI NIH HHS U01 HG010219NHGRI NIH HHS U01 HG010230NHGRI NIH HHS U01 HG010233NINDS NIH HHS U54 NS093793NINDS NIH HHS U54 NS108251Wellcome Trust
6 · The paper itself

Abstract

There are over 7000 rare diseases, some affecting 3500 or fewer patients in the United States. Due to clinicians' limited experience with such diseases and the heterogeneity of clinical presentations, ~70% of individuals seeking a diagnosis remain undiagnosed. Deep learning has demonstrated success in aiding the diagnosis of common diseases. However, existing approaches require labeled datasets with thousands of diagnosed patients per disease. We present SHEPHERD, a few-shot learning approach for multi-faceted rare disease diagnosis. SHEPHERD performs deep learning over a knowledge graph enriched with rare disease information and is trained on a dataset of simulated rare disease patients. We demonstrate SHEPHERD's effectiveness across diverse diagnostic tasks, performing causal gene discovery, retrieving "patients-like-me", and characterizing novel disease presentations, using real-world cohorts from the Undiagnosed Diseases Network (N = 465), MyGene2 (N = 146), and the Deciphering Developmental Disorders study (N = 1431). SHEPHERD demonstrates the potential of knowledge-grounded deep learning to accelerate rare disease diagnosis.

Identifiers

PMID40542121
PMCPMC12181314

What Socratic holds

Textmetadata
LicenceCC BY-NC-ND
Read underepoch 390

Registered trials

None linked

Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the Socratic graph.