Result

Adjusted rand index of scGPT on Aorta single-cell dataset

0.47 Adjusted rand index

Methods

Tested configuration: scGPT
Task: Cell-type structure in frozen embeddings
Dataset: Aorta single-cell dataset
Related family profiles: scGPT
Procedure: k-means on pretrained cell embeddings; agreement with original cell-type labels.
Evaluation: scGPT: Cell-type structure in frozen embeddings
Coverage: Not reported scored / Not reported eligible
Uncertainty: Not yet extracted
Evidence: Independent external evaluation · source checkedGenePT: A Simple But Effective Foundation Model for Genes and Cells Built From ChatGPT · Table 2, Aorta / Cell type row, scGPT ARI column

A source-checked result verifies the numerical transcription, not every model or protocol detail. Evaluation metadata: needs review. Source checked does not mean independently reproduced.

Reproduction

Split: Not reported
Adaptation: Not reported
Scoring implementation: Not reported

No execution recipe has been verified for this exact configuration and evaluation. A benchmark's general instructions may use different inputs, splits or model settings.

Reproducing this published result requires matching its model configuration, data, split and scorer. Source checking or a successful smoke test does not establish score reproduction.

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

1 evidence row matching the loaded filters

Claims, original sources and review scope · Release 2026-11-02-a7f03c9e55a5
Property and statement	Original source and location	Review and provenance
Printed value 0.47 Individual claims	GenePT: A Simple But Effective Foundation Model for Genes and Cells Built From ChatGPT Original source ↗ Table 2, Aorta / Cell type row, scGPT ARI column Version: PMC archival version PMC10614824.2 Retrieved: 2026-09-16T10:44:03.399274+00:00	source checked ["ai-assisted-source-review"] · 2026-09-16T10:44:03.402215+00:00 independent paper Audit details First Cell type row belongs to Aorta, not preceding Phenotype row or later organs. ARI is first in each three-metric method block, not AMI/ASW. Source check verifies central value and table context, not experiment reproduction or all metadata. Field: `attributes.printed_value` Claim: claim-lit-b3-012 Source artifact SHA-256: `230a2ec55458d9243eaeeebf3244df7409eb02d47f4b809ee56a06dcb6fdd047` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact Extraction artifact SHA-256: `230a2ec55458d9243eaeeebf3244df7409eb02d47f4b809ee56a06dcb6fdd047` Extraction artifact

How to read the evidence

Sources and history

Release 2026-11-02-a7f03c9e55a5 · Record review: source checked

1 source record and release history

GenePT: A Simple But Effective Foundation Model for Genes and Cells Built From ChatGPT · Original source · PMC archival version PMC10614824.2

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: lit-b3-012

areas: cells-tissues
tasks: Cell-type structure in frozen embeddings
printed value: 0.47
numeric value: 0.47
metric: adjusted-rand-index
metric direction: unknown
unit: unitless
source locator: Table 2, Aorta / Cell type row, scGPT ARI column
review: method: ai-assisted-source-review; reviewer: codex; reviewed at: 2026-09-16T10:44:03.402215+00:00; evidence: {"table_xml_id": "T6", "row_cells": ["", "Cell type", "0.21", "0.31", "−0.04", "0.47", "0.64", "0.18", "0.54", "0.60", "0.03", "0.31", "0.47", "0.04"], "selected_cell_zero_based": 5, "selected_cell_xml": "<td align=\"center\" valign=\"top\" rowspan=\"1\" colspan=\"1\">0.47</td>", "caption": "Assessing the Association Between Different Latent Cell Representations and Biological Annotations.This analysis involves datasets representing cells from circulatory systems (Aorta and Artery), bone tissues (Bones, Myeloid), the Pancreas, and immune cells collected from healthy individuals and patients with Multiple Sclerosis. We utilized pretrained Geneformer and scGPT embeddings for this task. The Adjusted Rand Index (ARI) and Adjusted Mutual Information (AMI) were computed to compare the labels derived from k-means clustering with the true annotations of the original samples (higher values indicate better alignment); the Average Silhouette Width (ASW) was calculated using the true annotations of original samples to assess the cohesion and separation of the clusters."}; artifact sha256: 230a2ec55458d9243eaeeebf3244df7409eb02d47f4b809ee56a06dcb6fdd047; retrieval url: https://www.ebi.ac.uk/europepmc/webservices/rest/PMC10614824/fullTextXML; method note: independent; table-level review; reviewer note: role: omics research agent; independent source-table review; not human review; note: First Cell type row belongs to Aorta, not preceding Phenotype row or later organs. ARI is first in each three-metric method block, not AMI/ASW. Source check verifies central value and table context, not experiment reproduction or all metadata.
legacy id: lit-b3-012
missing metadata: dataset version: reason: unextracted; model version: reason: unextracted; split: reason: unextracted; uncertainty: reason: unextracted

Related records

evaluation: scGPT: Cell-type structure in frozen embeddings
subject: Reported Adjusted Rand Index for scGPT

Suggest a correction