LAB-Bench SeqQA — ORF amino-acid position
Finds the amino acid encoded at a specified position in the longest ORF of a DNA sequence.
1 evaluation run(s)
track · audited · verified 2026-07-21
LAB-Bench SeqQA
Sequence-comprehension and manipulation category spanning 15 formal tasks involving PCR, restriction digestion, ORFs, translation, GC content, and DNA–protein relationships.
Benchmark definition
What is counted
- Version
- repository-998a8e0
- Total
- 750 (questions across formal child tasks)
- Task formats
- multiple choice
- Capabilities
- Modalities
Version history
| Version | Status | Release / as-of | Total | Formal tracks |
|---|---|---|---|---|
paper-v3lab-bench-seqqa-paper-v3 | active | 2024-07-17 | 750 (questions across formal child tasks) | lab-bench-seqqa-orf-seq-aaid, lab-bench-seqqa-orf-seq-aaseq, lab-bench-seqqa-orf-seq-numlen, lab-bench-seqqa-orf-transeff, lab-bench-seqqa-pcr-gene-enzprimers, lab-bench-seqqa-pcr-gene-gibshindprimers, lab-bench-seqqa-pcr-gene-gibssmaprimers, lab-bench-seqqa-pcr-geneprimers-enz, lab-bench-seqqa-pcr-len-primers, lab-bench-seqqa-pcr-primers-len, lab-bench-seqqa-pcr-seq-enzprimers, lab-bench-seqqa-pcr-seq-primers, lab-bench-seqqa-prop-seq-gcpercent, lab-bench-seqqa-re-seq-lenfrags, lab-bench-seqqa-re-seq-numfrags |
repository-998a8e0lab-bench-seqqa-repository-998a8e0 | current | 2025-09-27 | 750 (questions across formal child tasks) | lab-bench-seqqa-orf-seq-aaid, lab-bench-seqqa-orf-seq-aaseq, lab-bench-seqqa-orf-seq-numlen, lab-bench-seqqa-orf-transeff, lab-bench-seqqa-pcr-gene-enzprimers, lab-bench-seqqa-pcr-gene-gibshindprimers, lab-bench-seqqa-pcr-gene-gibssmaprimers, lab-bench-seqqa-pcr-geneprimers-enz, lab-bench-seqqa-pcr-len-primers, lab-bench-seqqa-pcr-primers-len, lab-bench-seqqa-pcr-seq-enzprimers, lab-bench-seqqa-pcr-seq-primers, lab-bench-seqqa-prop-seq-gcpercent, lab-bench-seqqa-re-seq-lenfrags, lab-bench-seqqa-re-seq-numfrags |
Tracks and subsets
| ID | Count | Basis | Partition? | Notes |
|---|---|---|---|---|
ORF amino-acid positionlab-bench-seqqa-orf-seq-aaid | 50 | questions | Exclusive & exhaustive | — |
ORF amino-acid sequencelab-bench-seqqa-orf-seq-aaseq | 50 | questions | Exclusive & exhaustive | — |
ORF count above lengthlab-bench-seqqa-orf-seq-numlen | 50 | questions | Exclusive & exhaustive | — |
Translation efficiencylab-bench-seqqa-orf-transeff | 50 | questions | Exclusive & exhaustive | — |
Gene-to-restriction primerslab-bench-seqqa-pcr-gene-enzprimers | 50 | questions | Exclusive & exhaustive | — |
Gene-to-Gibson primers (HindIII)lab-bench-seqqa-pcr-gene-gibshindprimers | 50 | questions | Exclusive & exhaustive | — |
Gene-to-Gibson primers (SmaI)lab-bench-seqqa-pcr-gene-gibssmaprimers | 50 | questions | Exclusive & exhaustive | — |
Primers-to-restriction enzymeslab-bench-seqqa-pcr-geneprimers-enz | 50 | questions | Exclusive & exhaustive | — |
Amplicon length to primerslab-bench-seqqa-pcr-len-primers | 50 | questions | Exclusive & exhaustive | — |
Primers to amplicon lengthlab-bench-seqqa-pcr-primers-len | 50 | questions | Exclusive & exhaustive | — |
Sequence-to-restriction primerslab-bench-seqqa-pcr-seq-enzprimers | 50 | questions | Exclusive & exhaustive | — |
Amplicon sequence to primerslab-bench-seqqa-pcr-seq-primers | 50 | questions | Exclusive & exhaustive | — |
GC percentagelab-bench-seqqa-prop-seq-gcpercent | 50 | questions | Exclusive & exhaustive | — |
Restriction-fragment lengthslab-bench-seqqa-re-seq-lenfrags | 50 | questions | Exclusive & exhaustive | — |
Restriction-fragment countlab-bench-seqqa-re-seq-numfrags | 50 | questions | Exclusive & exhaustive | — |
Registered child tracks
15 normalized runs are registered across these 15 tracks. Open a track for its exact split, prompt group, metrics, results, and comparison charts.
LAB-Bench SeqQA — ORF amino-acid sequence
Translates the longest ORF in a DNA sequence to its amino-acid sequence.
1 evaluation run(s)
LAB-Bench SeqQA — ORF count above length
Counts open reading frames encoding proteins above a specified amino-acid length.
1 evaluation run(s)
LAB-Bench SeqQA — Translation efficiency
Selects an RNA sequence whose ORF context is most likely to yield high translation efficiency.
1 evaluation run(s)
LAB-Bench SeqQA — Gene-to-restriction primers
Selects restriction-cloning primers from a named gene and enzyme pair.
1 evaluation run(s)
LAB-Bench SeqQA — Gene-to-Gibson primers (HindIII)
Selects primers for Gibson assembly into a HindIII-linearized vector.
1 evaluation run(s)
LAB-Bench SeqQA — Gene-to-Gibson primers (SmaI)
Selects primers for Gibson assembly into a SmaI-linearized vector.
1 evaluation run(s)
LAB-Bench SeqQA — Primers-to-restriction enzymes
Infers restriction enzymes from a gene name and primer pair.
1 evaluation run(s)
LAB-Bench SeqQA — Amplicon length to primers
Selects primers that produce a requested amplicon length from a DNA template.
1 evaluation run(s)
LAB-Bench SeqQA — Primers to amplicon length
Calculates expected amplicon length from a primer pair and DNA template.
1 evaluation run(s)
LAB-Bench SeqQA — Sequence-to-restriction primers
Selects restriction-cloning primers from an explicit gene sequence and enzyme pair.
1 evaluation run(s)
LAB-Bench SeqQA — Amplicon sequence to primers
Selects primers that produce a requested amplicon sequence from a DNA template.
1 evaluation run(s)
LAB-Bench SeqQA — GC percentage
Calculates the rounded GC percentage of a DNA sequence.
1 evaluation run(s)
LAB-Bench SeqQA — Restriction-fragment lengths
Calculates fragment lengths after restriction digestion of a DNA sequence.
1 evaluation run(s)
LAB-Bench SeqQA — Restriction-fragment count
Calculates the number of fragments after restriction digestion of a DNA sequence.
1 evaluation run(s)
Scientific Task Atlas
Scientific task classification
partial for repository-998a8e0. Formal child tracks are mapped where they express a controlled Scientific Task; generic sequence arithmetic remains unclassified.
| Scientific task | Coverage | Count | Mapping | Evidence |
|---|---|---|---|---|
| DNA sequence design | explicitly-in-scope | Not reported Primer-design questions across formal SeqQA child tracks. | official-track high confidence | lab-bench-seqqa-evidence-paperA cross-track subtotal is not reported by the creator. |
| RNA processing and translation prediction | explicitly-in-scope | 200 questions Four 50-question ORF and translation formal child tracks. | official-track high confidence | lab-bench-seqqa-evidence-paperThe count is derived only from formal child-track sizes documented by the official release. |
Scientific coverage notes
| Domain | Coverage | Count | Interpretation |
|---|---|---|---|
| Protein sequence | explicitly-in-scope | Not reported | ORF tasks connect DNA/RNA and amino-acid outputs, but no official protein-only question count is published. |
| Protein design | not-in-scope | 0 | Primer design is included; protein sequence design is not. |
Evaluation registry
Works and run settings
A setting change—scope, prompt, tools, budget, grader, or repeats—creates a separate run. Charts never cross a comparability group.
Evaluated models / systems: Claude Opus 4, Claude Opus 4.1, Claude Sonnet 4, Claude Sonnet 4.5
Scopetrack
Shots10
Turnssingle-turn
System prompt publicNot reported
Reasoning / effortNot reported
BrowserNo
InternetNo
DatabasesNo
Code executionNo
ContainerNo
External toolsNo
Token budgetNot reported
Time / cost budgetNot reported
TemperatureNot reported
SeedNot reported
RepeatsNot reported
GraderNot reported
Statisticsreported point estimate with plotted error bars
ContaminationNot reported
Metrics, results, and full protocol
Metrics
| Metric | Kind / baseline | Unit | Aggregation | Threshold / tolerance |
|---|---|---|---|---|
| LAB-Bench score | absolute | proportion | Not reported | Not reported |
Results
| Model | Metric | Value | n |
|---|---|---|---|
| Claude Sonnet 4 | LAB-Bench score | 0.682 proportion Point label transcribed from Figure 9.2.4.4.A; numeric error-bar bounds are not reported. | Not reported |
| Claude Opus 4 | LAB-Bench score | 0.723 proportion Point label transcribed from Figure 9.2.4.4.A; numeric error-bar bounds are not reported. | Not reported |
| Claude Opus 4.1 | LAB-Bench score | 0.785 proportion Point label transcribed from Figure 9.2.4.4.A; numeric error-bar bounds are not reported. | Not reported |
| Claude Sonnet 4.5 | LAB-Bench score | 0.78 proportion Point label transcribed from Figure 9.2.4.4.A; numeric error-bar bounds are not reported. | Not reported |
Evidence
- section: Section 9.2.4.4, printed pages 132–133 (Four selected tracks, 10-shot prompting, no search/bioinformatics tools, threshold, and unreported scope/repeats/grader.) — supports /benchmark_version, /model_ids, /scope, /protocol, /metrics
- figure: Figure 9.2.4.4.A, printed page 133 (Exact point labels. Protocol Sonnet 4/4.5 labels are represented in the Claude for Life Sciences run to avoid duplicate result rows.) — supports /results
Comparable result views
LAB-Bench score
lab-bench-seqqa-anthropic-sonnet45-system-card · lab-bench-seqqa-anthropic-10shot-no-tools
Accessible data table
| Model | Value | Comparability group |
|---|---|---|
| Claude Sonnet 4 | 0.682 | lab-bench-seqqa-anthropic-10shot-no-tools |
| Claude Opus 4 | 0.723 | lab-bench-seqqa-anthropic-10shot-no-tools |
| Claude Opus 4.1 | 0.785 | lab-bench-seqqa-anthropic-10shot-no-tools |
| Claude Sonnet 4.5 | 0.78 | lab-bench-seqqa-anthropic-10shot-no-tools |
Evidence and change history
Source locators remain visible; expand an item to inspect the exact Registry fields it supports.
LAB-Bench: Measuring Capabilities of Language Models for Biology Research · table: Table 1 and Appendix Table 6; task-specific Appendix C sections (Category total, task definitions, domains, modalities, and creator evaluation.) · Supports 17 fields
/name/organizations/release_date/kind/domains/capabilities/modalities/task_counts/total/task_counts/basis/task_counts/subsets/access/level/access/license/versions/0/task_counts/total/versions/0/task_counts/basis/versions/0/task_counts/subsets/scientific_task_classification/entries/0/scientific_task_classification/entries/1
lab-bench-seqqa-repository-resource · repository-path: SeqQA/*-splits.json and task.py at 998a8e0a40cf116c80e1b0e7a805ebb5fb9fa838 (Public/private/total 600/150/750; exact child files and evaluator.) · Supports 13 fields
/latest_version/task_counts/total/task_counts/basis/task_counts/subsets/access/level/access/tasks/access/artifacts/access/grader/access/license/resources/versions/1/task_counts/total/versions/1/task_counts/basis/versions/1/task_counts/subsets