Skills › Scientific Agent Skills › primer-design › sources.md
sources.md
74 lines · skills/primer-design/references/sources.md at 5f2bc03
1# Sources and maintenance notes
3Checked 2026-10-01. Links below are upstream documentation or primary publications.
4Their availability does not imply that remote services, specialized assays, or
5every API option were exercised by the bundled test suite.
Sources and maintenance notes
Checked 2026-10-01. Links below are upstream documentation or primary publications. Their availability does not imply that remote services, specialized assays, or every API option were exercised by the bundled test suite.
Raw lines 1 to 67## Executable design and analysis
9- [Primer3 manual](https://primer3.org/manual.html): authoritative Boulder-IO design
10 tags, coordinates, constraint semantics, and output definitions. Distinguish
11 template binding length from products that include overhangs.
12- [primer3-py documentation](https://libnano.github.io/primer3-py/): documentation
13 currently identifies release 2.3.1 and links the Primer3 2.6.1 manual. Package and
14 embedded engine versions are distinct; record both at runtime.
15- [Primary Python API](https://libnano.github.io/primer3-py/api/bindings.html):
16 `design_primers`, Tm/structure functions, physical parameters, and length limits.
17- [Thermodynamic result API](https://libnano.github.io/primer3-py/api/thermoanalysis.html):
18 result units, structure flags, and `get_libprimer3_version`.
19- [PyPI release metadata](https://pypi.org/pypi/primer3-py/json): current package
20 release remains 2.3.1; the bundled pin is unchanged.
Executable design and analysis
- Primer3 manual: authoritative Boulder-IO design tags, coordinates, constraint semantics, and output definitions. Distinguish template binding length from products that include overhangs.
- primer3-py documentation: documentation currently identifies release 2.3.1 and links the Primer3 2.6.1 manual. Package and embedded engine versions are distinct; record both at runtime.
- Primary Python API:
design_primers, Tm/structure functions, physical parameters, and length limits. - Thermodynamic result API: result units, structure flags, and
get_libprimer3_version. - PyPI release metadata: current package release remains 2.3.1; the bundled pin is unchanged.
22## Reference specificity
24- [NCBI Primer-BLAST](https://www.ncbi.nlm.nih.gov/tools/primer-blast/): current
25 assay-design/specificity interface and field-level help.
26- [NCBI search tips](https://www.ncbi.nlm.nih.gov/tools/primer-blast/search_tips.html):
27 accession-based inputs, target assignment, and search interpretation.
28- [Ye et al. (2012), Primer-BLAST](https://pmc.ncbi.nlm.nih.gov/articles/PMC3412702/):
29 combined design, alignment, and amplicon-level specificity workflow.
30- [BLAST+ nucleotide options](https://www.ncbi.nlm.nih.gov/sites/books/NBK279684/table/appendices.T.blastn_application_options/):
31 short-query tasks and search settings. Consult the installed executable's help
32 because the Bookshelf table also contains historical release information.
34The local Python contract remains `design_primers(seq_args, global_args)` returning
35a result dictionary, and snake_case thermodynamic functions returning a float Tm
36or `ThermoResult`. BLAST screening invokes local `makeblastdb` and `blastn`, parsing
37declared tabular fields; it has no network API/authentication requirement. The
38Primer-BLAST website was read for database and mismatch-field semantics; no public
39sequence was submitted and no undocumented request/response endpoint is supported.
Reference specificity
- NCBI Primer-BLAST: current assay-design/specificity interface and field-level help.
- NCBI search tips: accession-based inputs, target assignment, and search interpretation.
- Ye et al. (2012), Primer-BLAST: combined design, alignment, and amplicon-level specificity workflow.
- BLAST+ nucleotide options: short-query tasks and search settings. Consult the installed executable's help because the Bookshelf table also contains historical release information.
The local Python contract remains design_primers(seq_args, global_args) returning a result dictionary, and snake_case thermodynamic functions returning a float Tm or ThermoResult. BLAST screening invokes local makeblastdb and blastn, parsing declared tabular fields; it has no network API/authentication requirement. The Primer-BLAST website was read for database and mismatch-field semantics; no public sequence was submitted and no undocumented request/response endpoint is supported.
41## Thermodynamic foundations
43- [SantaLucia (1998)](https://pubmed.ncbi.nlm.nih.gov/9465037/): DNA nearest-neighbor
44 parameter framework, DOI `10.1073/pnas.95.4.1460`.
45- [Owczarzy et al. (2008)](https://pubmed.ncbi.nlm.nih.gov/18422348/): mixed-ion
46 corrections, DOI `10.1021/bi702363u`.
47- [Owczarzy et al. (2011)](https://pubmed.ncbi.nlm.nih.gov/21928795/): LNA-specific
48 thermodynamics and mismatch behavior, DOI `10.1021/bi200904e`.
Thermodynamic foundations
- SantaLucia (1998): DNA nearest-neighbor parameter framework, DOI
10.1073/pnas.95.4.1460. - Owczarzy et al. (2008): mixed-ion corrections, DOI
10.1021/bi702363u. - Owczarzy et al. (2011): LNA-specific thermodynamics and mismatch behavior, DOI
10.1021/bi200904e.
50## Assay-specific evidence
52- [Bustin et al. (2025), MIQE 2.0](https://academic.oup.com/clinchem/article/71/6/634/8119148):
53 current qPCR assay reporting and validation guidance,
54 DOI `10.1093/clinchem/hvaf043`.
55- [NEB overlap-assembly primer guidance](https://www.neb.com/en-sg/tools-and-resources/video-library/primer-design-and-fragment-assembly-using-gibson-assembly?autoplay=1):
56 connecting annealing regions to assembly overlaps.
57- [Brown et al., PrimerPooler](https://pubmed.ncbi.nlm.nih.gov/32161789/): multiplex
58 pool allocation; consult the linked erratum when reproducing published details.
59- [Rose et al. (2003), CODEHOP](https://pubmed.ncbi.nlm.nih.gov/12824413/):
60 protein-alignment-guided degenerate primers, DOI `10.1093/nar/gkg524`.
61- [Li and Dahiya (2002), MethPrimer](https://pubmed.ncbi.nlm.nih.gov/12424112/):
62 bisulfite-specific design constraints, DOI `10.1093/bioinformatics/18.11.1427`.
63- [Tusnády et al. (2005), BiSearch](https://pubmed.ncbi.nlm.nih.gov/15653630/):
64 converted-reference mispriming analysis, DOI `10.1093/nar/gni012`.
Assay-specific evidence
- Bustin et al. (2025), MIQE 2.0: current qPCR assay reporting and validation guidance, DOI
10.1093/clinchem/hvaf043. - NEB overlap-assembly primer guidance: connecting annealing regions to assembly overlaps.
- Brown et al., PrimerPooler: multiplex pool allocation; consult the linked erratum when reproducing published details.
- Rose et al. (2003), CODEHOP: protein-alignment-guided degenerate primers, DOI
10.1093/nar/gkg524. - Li and Dahiya (2002), MethPrimer: bisulfite-specific design constraints, DOI
10.1093/bioinformatics/18.11.1427. - Tusnády et al. (2005), BiSearch: converted-reference mispriming analysis, DOI
10.1093/nar/gni012.
66## When maintaining this skill
68Check upstream parameter names, supported alphabets/lengths, concentrations, output
69coordinates, and returned warnings before changing package pins. Re-run the
70repository suite and documented examples. Test off-target products and negative
71cases, not only whether a program returns primers. Preserve the distinction between
72locally tested numerical behavior, external service documentation, and empirical
73assay validation. A newer package or web interface does not justify upgrading old
74assay claims without rerunning their evidence.
When maintaining this skill
Check upstream parameter names, supported alphabets/lengths, concentrations, output coordinates, and returned warnings before changing package pins. Re-run the repository suite and documented examples. Test off-target products and negative cases, not only whether a program returns primers. Preserve the distinction between locally tested numerical behavior, external service documentation, and empirical assay validation. A newer package or web interface does not justify upgrading old assay claims without rerunning their evidence.
Raw lines 66 to 74