Evaluation

GPT-4 Zero shot on PAVS candidate gene sets

Published comparison; transcribed, not reproduced.

Research readiness

These checks assess whether the evidence supports a reproducible investigation. A source-checked score alone does not meet these requirements.

Release 2026-10-10-6e93f504adfc · Evidence verified: Not verified

Evidence incomplete

Replay metrics

Exact outcomes, predictions, identifiers and evaluator are connected.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
metric replay: verification is missing

Verified: Not verified

Evidence incomplete

Investigate discrepancies

Replay evidence includes annotations and an assessment of dependence. Unknown independence permits descriptive analysis only.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
metric replay: verification is missing
annotations: verification is missing
dependence: verification is missing

Verified: Not verified

Evidence incomplete

Run locally

A pinned recipe describes the inputs, environment and resource requirements.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
recipe pinned: verification is missing
resource estimate: verification is missing

Verified: Not verified

Evidence incomplete

Validate independently

Separate data and exposure records support an independent test.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
artifact hashes: verification is missing
join integrity: verification is missing
score semantics: verification is missing
independent validation: verification is missing
overlap checked: verification is missing

Verified: Not verified

Readiness describes the evidence in this release. Availability on your computer is checked separately when an investigation runs. Existing data exposure can prevent independent validation even when files are available.

Artifacts and reproduction

No verified artifact manifest is connected to this record yet. The gaps above identify what is needed before analysis can begin.

Read reviewed discrepancy investigations

Evaluation results

1 evaluation · 20 results. Different protocols are not a single leaderboard.

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.5 auprc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 100, 'AUPR', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.895 auroc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 100, 'ROC AUC', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	56.4% top-1-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 100, 'Hits@1 (%)', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	77.2% top-10-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 100, 'Hits@10 (%)', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.753 auprc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 25, 'AUPR', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.97 auroc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 25, 'ROC AUC', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	75.4% top-1-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 25, 'Hits@1 (%)', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	97.8% top-10-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 25, 'Hits@10 (%)', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.943 auprc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 5, 'AUPR', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.989 auroc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 5, 'ROC AUC', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	93% top-1-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 5, 'Hits@1 (%)', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	100% top-10-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 5, 'Hits@10 (%)', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.639 auprc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 50, 'AUPR', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.949 auroc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 50, 'ROC AUC', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	66.4% top-1-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 50, 'Hits@1 (%)', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	91% top-10-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 50, 'Hits@10 (%)', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.553 auprc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 75, 'AUPR', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	0.992 auroc fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 75, 'ROC AUC', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	59.8% top-1-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 75, 'Hits@1 (%)', column 'PAVS' 'GPT-4' 'Zero shot'
Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)	Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3) Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)	83.4% top-10-accuracy percent · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-4 Zero shot on PAVS candidate gene sets rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Aggregation: Not reported The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Table 3 row Size 75, 'Hits@10 (%)', column 'PAVS' 'GPT-4' 'Zero shot'

Source checking is not independent reproduction. Release 2026-10-10-6e93f504adfc.

Evaluation procedure

rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets

Configuration: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)
Protocol: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3)
Dataset: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)

origin: Author-reported evaluation
configuration: Primary source as retrieved 2026-10-09
dataset version: As published
split: No split
population: PAVS genotype-phenotype pairs
inputs: HPO-coded phenotypes and a candidate gene list
adaptation: Zero shot
metric implementation: Hits@k, ROC AUC and AUPR over candidate sets
aggregation: Per candidate-set size
budget: Candidate sets of 5, 25, 50, 75 and 100 genes
protocol id: rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets

Metadata review: source checked. Unreported conditions prevent automatic comparisons.

Reproduction

Split: No split
Adaptation: Zero shot
Scoring implementation: Hits@k, ROC AUC and AUPR over candidate sets

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.

19 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-10-6e93f504adfc
Property and statement	Original source and location	Review and provenance
attributes.comparison.adaptation Zero shot Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.adaptation` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.comparison.aggregation Per candidate-set size Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.aggregation` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.comparison.budget Candidate sets of 5, 25, 50, 75 and 100 genes Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.budget` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.comparison.dataset_version As published Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.dataset_version` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.comparison.inputs HPO-coded phenotypes and a candidate gene list Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.inputs` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.comparison.metric_implementation Hits@k, ROC AUC and AUPR over candidate sets Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.metric_implementation` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.comparison.population PAVS genotype-phenotype pairs Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.population` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.comparison.protocol_id rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.protocol_id` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.comparison.split No split Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.comparison.split` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
attributes.limitations 1 values The authors designed the GPT-4 prompts, including the one-shot chain-of-thought prompt Q4 (Table 1; Methods 'Prompt engineering'; Author contributions), so these rows are author_reported. Context-only references	The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients Original source ↗ Table 3 column 'PAVS', 'GPT-4' 'Zero shot' Version: Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML Retrieved: 2026-10-09T21:20:27Z	not individually reviewed No individual claim review recorded author reported Audit details Field: `attributes.limitations` Source artifact SHA-256: `75851f55d59fb0f194ca8bbcf10f47b6f593e8658239dc0fb85f5d2844236a3b` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact

All evidence (JSONL) (gzip)How to read the evidence

Sources and history

Release 2026-10-10-6e93f504adfc · Record review: source checked

1 source records and release history

The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients · Original source · Scientific Reports 15, published 2025-04-29; PMC12041562 full-text XML

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: rare-ranking-20261009-eval-kafkas2025-pavs-gpt-4-zero-shot

areas: dna-genomes
contexts: clinical_research
origin: author_reported
protocol: rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets
version: Primary source as retrieved 2026-10-09
comparison: dataset version: As published; split: No split; population: PAVS genotype-phenotype pairs; inputs: HPO-coded phenotypes and a candidate gene list; adaptation: Zero shot; metric implementation: Hits@k, ROC AUC and AUPR over candidate sets; aggregation: Per candidate-set size; budget: Candidate sets of 5, 25, 50, 75 and 100 genes; protocol id: rare-ranking-20261009-protocol-kafkas2025-pavs-gene-sets
source locator: Table 3 column 'PAVS', 'GPT-4' 'Zero shot'
limitations: The authors designed the GPT-4 prompts, including the one-shot chain-of-thought prompt Q4 (Table 1; Methods 'Prompt engineering'; Author contributions), so these rows are author_reported.

Related records

system: GPT-4 (gpt-4-1106-preview), zero-shot prompt (Kafkas et al. 2025)
assessment: Ranking the causative gene within synthetic candidate sets of 5 to 100 genes, PAVS (Kafkas et al. 2025 Table 3)
data: PAVS Saudi phenotype-associated variants, 500 genes (Kafkas et al. 2025)
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size100-auprc auprc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size100-auroc auroc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size100-hits1 top-1-accuracy
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size100-hits10 top-10-accuracy
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size25-auprc auprc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size25-auroc auroc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size25-hits1 top-1-accuracy
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size25-hits10 top-10-accuracy
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size5-auprc auprc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size5-auroc auroc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size5-hits1 top-1-accuracy
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size5-hits10 top-10-accuracy
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size50-auprc auprc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size50-auroc auroc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size50-hits1 top-1-accuracy
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size50-hits10 top-10-accuracy
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size75-auprc auprc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size75-auroc auroc
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size75-hits1 top-1-accuracy
evaluation: kafkas2025-pavs-gpt-4-zero-shot-size75-hits10 top-10-accuracy

Suggest a correction