Back to results

CompBioBench item

vcf-infer-ancestry-q1

Question

You are given vcf.infer.ancestry.q1.vcf.gz, a bgzipped VCF containing variants on chromosome 1 for four unrelated individuals: Sample1, Sample2, Sample3, Sample4. Infer each individual’s ancestry as one of: AFR (African), ASJ (Ashkenazi Jew), EUR (European), EAS (East Asian), MID (Middle Eastern), SAS (South Asian), OCE (Oceania). Output 4 labels in Sample1→Sample4 order, comma-separated with no spaces (e.g., AFR,EUR,SAS,OCE).

Predicted answer
AFR,AFR,SAS,EURMost likely answer for this question; not uniquely determined from submitted scores.
Domain
Population Genetics
Question style
Metadata Recovery

An answer can be uniquely inferred for 54 of 100 questions from submitted scores. Green: correct · Red: incorrect. For questions without a uniquely determined answer: Blue: matches prediction · Orange: differs from prediction. These are inferred answer checks, not official per-question grades.

Galaxy

12/12 terminal item results

UDT = user-defined tool: custom code executed by Galaxy. Route badges summarize analysis tools, excluding uploads and automatic conversions. How usage is counted.

Codex GPT-5.5

3 replicates
Codex GPT-5.5

Replicate 1

Matches predictionUDT-only
AnswerAFR,AFR,SAS,EUR
Workflow
BCFtoolsVCF ancestry inferenceVCF ancestry inference v2VCF ancestry inference v3VCF ancestry inference v44 more tools
Total tokens
4.3M
Output
39.8K
Codex GPT-5.5

Replicate 2

Matches predictionUDT-only
AnswerAFR,AFR,SAS,EUR
Workflow
VCF ancestry diagnosticVCF ancestry diagnostic gzip-awareVCF 1000 Genomes exact match diagnosticVCF 1000 Genomes exact match subsetVCF 1000 Genomes population summary
Total tokens
1.2M
Output
17.4K
Codex GPT-5.5

Replicate 3

RerunMatches predictionUDT-only
AnswerAFR,AFR,SAS,EUR
Workflow
VCF ancestry diagnosticgnomAD build allele diagnosticgnomAD ancestry classifiergnomAD parser debug1000G ancestry classifier1 more tools
Total tokens
3.3M
Output
26.4K

Codex GPT-5.6 Sol

3 replicates
Codex GPT-5.6 Sol

Replicate 1

Matches predictionUDT + standard tools
AnswerAFR,AFR,SAS,EUR
Workflow
Bcftools StatsGalaxy UDTancestry_qcAncestry QCAncestry VCF QC2 more tools
Total tokens
3.8M
Output
18.6K
Codex GPT-5.6 Sol

Replicate 2

Matches predictionUDT + standard tools
AnswerAFR,AFR,SAS,EUR
Workflow
Bcftools StatsVCF structure diagnosticSGDP exact genotype identity match
Total tokens
3.1M
Output
16.1K
Codex GPT-5.6 Sol

Replicate 3

Matches predictionUDT-only
AnswerAFR,AFR,SAS,EUR
Workflow
VCF ancestry diagnosticVCF ancestry diagnostic v2SGDP tree coordinate and allele overlap diagnosticSGDP tree coordinate and allele overlap diagnostic v2SGDP exact genotype ancestry match
Total tokens
3.6M
Output
17.4K

Codex + DeepSeek-v4-pro-0813

3 replicates
Codex + DeepSeek-v4-pro-0813

Replicate 1

Matches predictionUDT + standard tools
AnswerAFR,AFR,SAS,EUR
Workflow
Bcftools ViewAncestry Likelihood
Total tokens
51.1M
Output
92.0K
Codex + DeepSeek-v4-pro-0813

Replicate 2

Matches predictionUDT + standard tools
AnswerAFR,AFR,SAS,EUR
Workflow
Crossmap VcfAncestry ProbeAncestry Probe v2Ancestry Probe v3Ancestry Classify2 more tools
Total tokens
55.8M
Output
205.0K
Codex + DeepSeek-v4-pro-0813

Replicate 3

Matches predictionUDT-only
AnswerAFR,AFR,SAS,EUR
Workflow
SGDP ancestry match
Total tokens
42.2M
Output
74.8K

Codex GPT-5.6 Luna

3 replicates
Codex GPT-5.6 Luna

Replicate 1

Matches predictionUDT + standard tools
AnswerAFR,AFR,SAS,EUR
Workflow
Bcftools StatsBcftools QueryTp Head ToolVCF ancestry input diagnosticsVCF ancestry marker probe5 more tools
Total tokens
20.8M
Output
94.5K
Codex GPT-5.6 Luna

Replicate 2

Matches predictionUDT + standard tools
AnswerAFR,AFR,SAS,EUR
Workflow
Bcftools StatsBcftools QueryVCF ancestry diagnosticVCF pairwise ancestry diagnosticUpload16 more tools
Total tokens
18.5M
Output
76.9K
Codex GPT-5.6 Luna

Replicate 3

Matches predictionUDT-only
AnswerAFR,AFR,SAS,EUR
Workflow
VCF ancestry input diagnosticUpload1VCF reference convention probeVCF reference identity probeRemote indexed VCF probe2 more tools
Total tokens
13.4M
Output
50.8K

Anycode

13/13 terminal item results

Codex GPT-5.5

3 replicates
Codex GPT-5.5

Replicate 1

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsPythonPLINK
Codex GPT-5.5

Replicate 2

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsPythonPLINKSamtools
Codex GPT-5.5

Replicate 3

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsPLINKPythonpysam

Codex GPT-5.6 Sol

3 replicates
Codex GPT-5.6 Sol

Replicate 1

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsPLINKPython
Codex GPT-5.6 Sol

Replicate 2

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codePLINKPython
Codex GPT-5.6 Sol

Replicate 3

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsPLINKPython

Codex + DeepSeek-v4-pro-0813

3 replicates
Codex + DeepSeek-v4-pro-0813

Replicate 1

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codePythonBCFtoolsPLINKSciPy1 more tools
Total tokens
14.5M
Output
63.2K
Codex + DeepSeek-v4-pro-0813

Replicate 2

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsPLINKPythonSamtools
Total tokens
11.9M
Output
56.7K
Codex + DeepSeek-v4-pro-0813

Replicate 3

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsPLINKPythonSciPy1 more tools
Total tokens
5.3M
Output
40.7K

Codex GPT-5.6 Luna

3 replicates
Codex GPT-5.6 Luna

Replicate 1

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsPLINKPythonpysam
Codex GPT-5.6 Luna

Replicate 2

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsPLINKSamtoolsPython2 more tools
Codex GPT-5.6 Luna

Replicate 3

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codeBCFtoolsSamtoolsPLINKPython1 more tools

Codex GPT-6 Astra

1 replicate
Codex GPT-6 Astra

Replicate 1

Matches prediction
AnswerAFR,AFR,SAS,EUR
Workflow
Local codePythonBCFtoolsPLINKpysam1 more tools
Total tokens
783.2K
Output
8.4K