Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Design qPCR primers and TaqMan/molecular beacon probes using primer3-py. Configure probe Tm, primer-probe spacing, and hydrolysis probe constraints for real-time PCR assays. Use when designing qPCR primers and probes.
.claude/skills/bio-primer-design-qpcr-primers/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 103% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 53% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 95% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 26% | 0% |
<!--
#
#
-->
Design primers and internal probes for quantitative PCR using primer3-py.
pythonimport primer3 from Bio import SeqIO
pythonsequence = 'ATGCGTACGATCGATCGATCGATCGATCGATCGATCGATCGATCGATCGATCGATCGATCGATCG' * 3 result = primer3.design_primers( seq_args={'SEQUENCE_TEMPLATE': sequence}, global_args={ 'PRIMER_PICK_LEFT_PRIMER': 1, 'PRIMER_PICK_RIGHT_PRIMER': 1, 'PRIMER_PICK_INTERNAL_OLIGO': 1, # Design internal probe 'PRIMER_PRODUCT_SIZE_RANGE': [[70, 150]], # Short amplicons for qPCR 'PRIMER_OPT_TM': 60.0, 'PRIMER_MIN_TM': 58.0, 'PRIMER_MAX_TM': 62.0, 'PRIMER_INTERNAL_OPT_TM': 70.0, # Probe Tm ~10C higher 'PRIMER_INTERNAL_MIN_TM': 68.0, 'PRIMER_INTERNAL_MAX_TM': 72.0, 'PRIMER_INTERNAL_MIN_SIZE': 18, 'PRIMER_INTERNAL_OPT_SIZE': 25, 'PRIMER_INTERNAL_MAX_SIZE': 30, } )
pythonnum_returned = result['PRIMER_PAIR_NUM_RETURNED'] print(f'Found {num_returned} primer/probe sets') for i in range(num_returned): left = result[f'PRIMER_LEFT_{i}_SEQUENCE'] right = result[f'PRIMER_RIGHT_{i}_SEQUENCE'] probe = result[f'PRIMER_INTERNAL_{i}_SEQUENCE'] probe_tm = result[f'PRIMER_INTERNAL_{i}_TM'] left_tm = result[f'PRIMER_LEFT_{i}_TM'] right_tm = result[f'PRIMER_RIGHT_{i}_TM'] product_size = result[f'PRIMER_PAIR_{i}_PRODUCT_SIZE'] print(f'Set {i}:') print(f' Forward: {left} (Tm: {left_tm:.1f}C)') print(f' Reverse: {right} (Tm: {right_tm:.1f}C)') print(f' Probe: {probe} (Tm: {probe_tm:.1f}C)') print(f' Product: {product_size}bp')
pythonresult = primer3.design_primers( seq_args={ 'SEQUENCE_TEMPLATE': sequence, 'SEQUENCE_TARGET': [100, 30], # Target region for probe }, global_args={ 'PRIMER_PICK_INTERNAL_OLIGO': 1, 'PRIMER_PRODUCT_SIZE_RANGE': [[60, 100], [100, 150]], # Prefer short 'PRIMER_NUM_RETURN': 5, # Primer parameters 'PRIMER_OPT_SIZE': 20, 'PRIMER_MIN_SIZE': 18, 'PRIMER_MAX_SIZE': 25, 'PRIMER_OPT_TM': 60.0, 'PRIMER_MIN_TM': 58.0, 'PRIMER_MAX_TM': 62.0, 'PRIMER_OPT_GC_PERCENT': 50.0, 'PRIMER_MIN_GC': 35.0, 'PRIMER_MAX_GC': 65.0, # Probe parameters (TaqMan: Tm 8-10C higher than primers) 'PRIMER_INTERNAL_OPT_SIZE': 25, 'PRIMER_INTERNAL_MIN_SIZE': 18, 'PRIMER_INTERNAL_MAX_SIZE': 30, 'PRIMER_INTERNAL_OPT_TM': 70.0, 'PRIMER_INTERNAL_MIN_TM': 68.0, 'PRIMER_INTERNAL_MAX_TM': 72.0, 'PRIMER_INTERNAL_MIN_GC': 30.0, 'PRIMER_INTERNAL_MAX_GC': 70.0, # Avoid G at 5' end of probe (quenches FAM) 'PRIMER_INTERNAL_MAX_SELF_ANY': 8, } )
python# Additional considerations for TaqMan probes global_args = { 'PRIMER_PICK_INTERNAL_OLIGO': 1, 'PRIMER_PRODUCT_SIZE_RANGE': [[70, 150]], # Probe Tm should be 8-10C higher than primers 'PRIMER_OPT_TM': 60.0, 'PRIMER_INTERNAL_OPT_TM': 70.0, # Probe should be closer to forward primer 'PRIMER_INTERNAL_MIN_SIZE': 18, 'PRIMER_INTERNAL_MAX_SIZE': 30, # Avoid long poly-X runs in probe 'PRIMER_INTERNAL_MAX_POLY_X': 3, }
python# For SYBR Green, design primers without probe result = primer3.design_primers( seq_args={'SEQUENCE_TEMPLATE': sequence}, global_args={ 'PRIMER_PICK_LEFT_PRIMER': 1, 'PRIMER_PICK_RIGHT_PRIMER': 1, 'PRIMER_PICK_INTERNAL_OLIGO': 0, # No probe 'PRIMER_PRODUCT_SIZE_RANGE': [[70, 200]], # Short for qPCR 'PRIMER_OPT_TM': 60.0, 'PRIMER_MIN_TM': 58.0, 'PRIMER_MAX_TM': 62.0, 'PRIMER_MAX_SELF_ANY': 4, # Strict for SYBR specificity 'PRIMER_MAX_SELF_END': 2, 'PRIMER_PAIR_MAX_COMPL_ANY': 4, 'PRIMER_PAIR_MAX_COMPL_END': 2, } )
python# For cDNA-specific amplification, target exon junction # Mark the exon junction position exon_junction = 150 # Position where exons meet result = primer3.design_primers( seq_args={ 'SEQUENCE_TEMPLATE': sequence, 'SEQUENCE_OVERLAP_JUNCTION_LIST': [exon_junction], # Primer must span }, global_args={ 'PRIMER_PRODUCT_SIZE_RANGE': [[70, 150]], 'PRIMER_OPT_TM': 60.0, 'PRIMER_MIN_3_PRIME_OVERLAP_OF_JUNCTION': 4, # Min bases on each side } )
python# Design primers for multiple targets with compatible Tms targets = [ {'name': 'gene1', 'seq': sequence1, 'target': [100, 30]}, {'name': 'gene2', 'seq': sequence2, 'target': [150, 30]}, ] results = [] for target in targets: result = primer3.design_primers( seq_args={ 'SEQUENCE_TEMPLATE': target['seq'], 'SEQUENCE_ID': target['name'], 'SEQUENCE_TARGET': target['target'], }, global_args={ 'PRIMER_PICK_INTERNAL_OLIGO': 1, 'PRIMER_PRODUCT_SIZE_RANGE': [[70, 150]], 'PRIMER_OPT_TM': 60.0, # Same Tm for all 'PRIMER_MAX_TM': 61.0, 'PRIMER_MIN_TM': 59.0, 'PRIMER_INTERNAL_OPT_TM': 70.0, } ) results.append(result)
python# Verify Tm with primer3's thermodynamic calculations primer_seq = 'ATGCGATCGATCGATCGATC' # Standard Tm tm = primer3.calc_tm(primer_seq) print(f'Standard Tm: {tm:.1f}C') # Tm with specific salt conditions (match your qPCR master mix) tm_adjusted = primer3.calc_tm( primer_seq, mv_conc=50.0, # Monovalent cation (K+, Na+) mM dv_conc=3.0, # Divalent cation (Mg2+) mM dntp_conc=0.8, # dNTP mM (reduces free Mg2+) dna_conc=250.0, # Primer concentration nM ) print(f'Adjusted Tm: {tm_adjusted:.1f}C')
pythonimport pandas as pd def qpcr_results_to_df(result): rows = [] for i in range(result['PRIMER_PAIR_NUM_RETURNED']): row = { 'pair': i, 'forward': result[f'PRIMER_LEFT_{i}_SEQUENCE'], 'reverse': result[f'PRIMER_RIGHT_{i}_SEQUENCE'], 'fwd_tm': result[f'PRIMER_LEFT_{i}_TM'], 'rev_tm': result[f'PRIMER_RIGHT_{i}_TM'], 'product_size': result[f'PRIMER_PAIR_{i}_PRODUCT_SIZE'], } if f'PRIMER_INTERNAL_{i}_SEQUENCE' in result: row['probe'] = result[f'PRIMER_INTERNAL_{i}_SEQUENCE'] row['probe_tm'] = result[f'PRIMER_INTERNAL_{i}_TM'] rows.append(row) return pd.DataFrame(rows) df = qpcr_results_to_df(result) print(df)
| Parameter | Primers | TaqMan Probe | |-----------|---------|--------------| | Length | 18-25 bp | 18-30 bp | | Tm | 58-62C | 68-72C | | GC% | 35-65% | 30-70% | | Amplicon | 70-150 bp | - | | 5' base | Any | Avoid G (quenches FAM) |
<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 13,198 | 12,946 | -2% | 1 | 1 | 0% | 2,644 | 5,633 | +113% | 0 | 0 | — |
case-02 | pass→pass | 14,886 | 15,404 | +3% | 1 | 1 | 0% | 3,276 | 5,893 | +80% | 0 | 0 | — |
case-03 | fail→fail | 13,313 | 11,324 | -15% | 1 | 1 | 0% | 2,556 | 5,220 | +104% | 0 | 0 | — |
case-04 | fail→pass | 11,735 | 9,933 | -15% | 1 | 1 | 0% | 2,453 | 4,968 | +103% | 0 | 0 | — |
case-05 | fail→pass | 11,734 | 5,435 | -54% | 1 | 1 | 0% | 2,531 | 3,874 | +53% | 0 | 0 | — |
case-06 | pass→pass | 13,843 | 6,644 | -52% | 1 | 1 | 0% | 2,648 | 4,149 | +57% | 0 | 0 | — |
case-07 | pass→pass | 8,563 | 3,750 | -56% | 1 | 1 | 0% | 1,633 | 3,471 | +113% | 0 | 0 | — |
case-08 | pass→pass | 9,794 | 12,127 | +24% | 1 | 1 | 0% | 2,265 | 4,452 | +97% | 0 | 0 | — |
case-09 | pass→pass | 6,622 | 4,371 | -34% | 1 | 1 | 0% | 1,405 | 3,566 | +154% | 0 | 0 | — |
case-10 | pass→pass | 16,674 | 7,189 | -57% | 1 | 1 | 0% | 3,259 | 4,277 | +31% | 0 | 0 | — |
case-11 | fail→pass | 32,751 | 33,501 | +2% | 1 | 1 | 0% | 4,112 | 6,315 | +54% | 0 | 0 | — |
case-12 | fail→pass | 9,176 | 3,284 | -64% | 1 | 1 | 0% | 1,763 | 3,434 | +95% | 0 | 0 | — |
case-13 | pass→pass | 5,305 | 3,500 | -34% | 1 | 1 | 0% | 1,054 | 3,533 | +235% | 0 | 0 | — |
case-14 | fail→pass | 15,997 | 5,478 | -66% | 1 | 1 | 0% | 3,145 | 3,970 | +26% | 0 | 0 | — |
case-15 | pass→pass | 7,045 | 3,420 | -51% | 1 | 1 | 0% | 1,434 | 3,475 | +142% | 0 | 0 | — |
case-16 | fail→pass | 9,712 | 3,576 | -63% | 1 | 1 | 0% | 1,932 | 3,540 | +83% | 0 | 0 | — |
case-17 | pass→pass | 3,543 | 1,663 | -53% | 1 | 1 | 0% | 620 | 3,093 | +399% | 0 | 0 | — |
case-18 | fail→pass | 8,173 | 2,713 | -67% | 1 | 1 | 0% | 1,318 | 3,277 | +149% | 0 | 0 | — |
case-19 | fail→pass | 22,937 | 3,536 | -85% | 1 | 1 | 0% | 1,872 | 3,497 | +87% | 0 | 0 | — |
case-20 | pass→pass | 10,894 | 11,079 | +2% | 1 | 1 | 0% | 2,029 | 5,016 | +147% | 0 | 0 | — |
case-21 | pass→pass | 13,646 | 11,050 | -19% | 1 | 1 | 0% | 3,054 | 5,515 | +81% | 0 | 0 | — |
case-22 | pass→pass | 19,026 | 12,457 | -35% | 1 | 1 | 0% | 3,904 | 5,130 | +31% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +36 percentage points is the difference between those two pass rates over the 21 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 7/26/2026 | +36% |
Other measured skills in the registry, with their headline benchmark lift.