1What the regulations expect of an instrument interface.
Neither US CGMP nor EU GMP Annex 11 names a LIMS, but their records rules decide what an integration has to carry. In the US, laboratory records must include “a complete record of all data secured in the course of each test, including all graphs, charts, and spectra from laboratory instrumentation”, identified to the material and lot tested, with every calculation, the analyst’s initials or signature, and those of a second person who reviewed the original records.¹ Computers may be used, but input to and output from them must be checked for accuracy, with the degree and frequency of checking based on the system’s complexity and reliability, and backup data must be “exact and complete” and secure from alteration, inadvertent erasure or loss.²
EU GMP Annex 11 asks that systems exchanging data electronically “include appropriate built-in checks for the correct and secure entry and processing of data”, that critical systems have a description of their data flows and interfaces, and that critical data entered by hand get an additional accuracy check by a second operator or validated electronic means.³ The PIC/S data integrity guidance, PI 041-1 of July 2021, adds that interfaces should be assessed during validation “to ensure the correct and complete transfer of data”, can be verified with secure transfer, encryption or checksums, and where applicable should be qualified to transfer GMP data automatically.⁴
| Expectation | Source |
|---|---|
| Complete data from each test, including instrument graphs, charts and spectra | 21 CFR 211.194(a)(4) |
| Calculations recorded, and original records reviewed by a second person | 21 CFR 211.194(a)(5) and (a)(8) |
| Input and output checked for accuracy; exact, complete and secure backups | 21 CFR 211.68(b) |
| Built-in checks on data exchanged between systems | EU GMP Annex 11, section 5 |
| A second check on critical data entered by hand | EU GMP Annex 11, section 6; PIC/S PI 041-1, section 9.7 |
| Interfaces validated for correct and complete transfer | PIC/S PI 041-1, section 9.4 |
| System clocks synchronised with connected systems | PIC/S PI 041-1, section 9.5 |
2Why the reported result is not the whole record.
A LIMS usually wants one number per test. The record the regulations protect is larger. FDA says a printed chromatogram is neither a true copy of the electronic raw data under 21 CFR 211.180(d) nor an exact and complete copy under 211.68, because it “does not generally include, for example, the injection sequence, instrument method, integration method, or the audit trail”. The electronic records “must be maintained” and be readily available to QA or an FDA investigator.⁵
FDA’s data integrity guidance of December 2018 draws the line by record type. Balances and pH meters may create a paper printout as the original record, which must then be kept. Records from instruments such as FT-IR, “whether stand-alone or networked”, are dynamic and can be reprocessed, so a printout does not preserve the complete original record. If chromatography is reprocessed, each result is kept for review; saving only the final result is not acceptable.⁶ PIC/S says complete capture of raw data is normally needed to reconstruct an analysis, along with relevant metadata such as user identity, times and units of measure.⁴
So a sound integration carries the reportable result into the LIMS and keeps a link from that result to the original electronic record, wherever that record is stored.
3Choosing a route for each instrument.
| Route | Typical use | What the guidance expects |
|---|---|---|
| Direct printout | Simple instruments that store no data, such as balances and pH meters | The printout is the original record: signed and dated by the person who made it, marked with the sample ID or batch, and attached to the testing record |
| Manual entry | Values read from a display or a printout | A second operator or a validated electronic check on critical data, with the original kept after transcription |
| File export from the instrument computer | Instrument software that writes result and raw data files | Data sent directly to secure storage rather than copied from a local drive where it could be altered, and controls on any temporary local storage |
| Automated interface | Networked instruments and data systems | The interface between source, acquisition and recording systems validated, with checks on the completeness of data and metadata |
The routes and their controls come from PIC/S PI 041-1, sections 8.9, 9.4, 9.7 and 9.10, and Annex 11, section 6.⁴³ PIC/S is plain about the ranking: “Well designed and qualified automated data transfer is much more reliable than any manual data transfer conducted by humans.” It discourages hybrid systems of paper and electronic records, saying they should be replaced whenever possible, and singles out standalone instruments, where data held on the instrument computer before transfer can be deleted or manipulated.⁴
4Validating the integration.
Qualifying the LIMS is not the same as validating what it does with instrument data. FDA treats each CGMP workflow as an intended use to be checked through validation, with the extent of validation commensurate with risk, and notes that qualifying a platform does not show that a record it generates contains the correct calculations.⁶ PIC/S warns that validating a system in isolation may not lower data integrity risk if a user can still influence which data is reported, and prefers a fully automated, validated process that keeps human intervention to a minimum.⁴
A practical way in is the data flow map PIC/S suggests for interfaced systems: follow one result from sample preparation and the weights entered, through acquisition and processing, to the reported value.⁴ Then test each hand-off for what can go wrong there: a failed or partial transfer, missing metadata, a result attached to the wrong sample, a clock out of step, a reprocessed result and a restore from backup. Annex 11 expects backups to be checked for integrity and restorability during validation and periodically, changes to be made under a defined procedure, and systems to be evaluated periodically to confirm they remain valid.³ The same thinking runs through audit trail review and ALCOA+.
5How it runs in Seal.
In Seal, instrument results arrive through the Seal IoT app: the vendor software saves an export, the Seal IoT app on a site computer collects it, and an upload Script maps it to sample and experiment records. Acquisition and evaluation stay in the instrument software, which remains the source of truth, and nothing is written back. Scientific data management keeps the original files, such as raw data, method, sequence, audit trail and report, attached to the sample, batch or investigation they support. An acquisition missing a required file stays incomplete, and reprocessing creates a derived record rather than overwriting the earlier one.
Where results come from a LIMS, the LIMS result integration guide sets out how to retain test identity, units, qualifiers and the source revision with each value, and Seal’s LabWare, LabVantage and SampleManager bindings read only. The chromatography data integration guide shows how to keep a repeated export, a revised analysis and a conflicting file distinguishable, and to preserve the source evaluation behind each value.
References
- 121 CFR 211.194, Laboratory records, eCFR. https:/
/ www. ecfr. gov/ current/ title- 21/ section- 211. 194 (read 6 November 2026) - 221 CFR 211.68, Automatic, mechanical, and electronic equipment, eCFR. https:/
/ www. ecfr. gov/ current/ title- 21/ section- 211. 68 (read 6 November 2026) - 3EudraLex Volume 4, Annex 11: Computerised Systems, revision 1, European Commission, January 2011. https:/
/ health. ec. europa. eu/ system/ files/ 2016- 11/ annex11_ 01- 2011_ en_ 0. pdf (read 6 November 2026) - 4PI 041-1, Good Practices for Data Management and Integrity in Regulated GMP/GDP Environments, PIC/S, 1 July 2021. https:/
/ picscheme. org/ docview/ 4234 (read 6 November 2026) - 5Questions and Answers on Current Good Manufacturing Practice Requirements: Records and Reports, FDA, updated 16 November 2022. https:/
/ www. fda. gov/ drugs/ guidances- drugs/ questions- and- answers- current- good- manufacturing- practice- requirements- records- and- reports (read 6 November 2026) - 6Data Integrity and Compliance With Drug CGMP: Questions and Answers, Guidance for Industry, FDA, December 2018. https:/
/ www. fda. gov/ media/ 119267/ download (read 6 November 2026)
AQuestions and answers
What is an SDMS, and how is it different from a LIMS?
A scientific data management system captures and keeps the files instruments produce, such as raw data, methods, sequences, audit trails and reports. A LIMS registers samples, runs tests against specifications and reports results. Where a lab uses both, the SDMS or the instrument’s own data system holds the original electronic record, and the LIMS holds the reportable result with a link back to it.
Can we keep printouts instead of the electronic records?
Only where the printout is the original record, as with a balance or pH meter that stores no data. For instruments whose records are dynamic, such as chromatography data systems or FT-IR, FDA says a printout does not preserve the complete original record, so the electronic record has to be kept and available for review.
Does every instrument interface need to be validated?
Every interface that carries GMP data does. FDA expects each CGMP workflow to be validated for its intended use, and EU and PIC/S guidance expect interfaces to be checked during validation for correct and complete transfer. How deeply you test follows the risk of the interface and the data it carries.
Is manual transcription of instrument results allowed?
Yes, with controls. Critical data typed in by hand needs a check by a second person or a validated electronic means, and the original record is kept after transcription. PIC/S says well designed and qualified automated transfer is much more reliable than manual transfer, so manual entry is best kept for the instruments that cannot be connected.
What are the risks of data sitting on the instrument computer?
Data held locally before transfer can be deleted or altered, especially on standalone instruments that are not networked. PIC/S expects data to go directly to secure storage rather than be copied later from a local drive, and expects the environment that first stores the data to have its own integrity controls.
Why do the instrument and LIMS clocks need to agree?
Times are part of the metadata needed to reconstruct an analysis. PIC/S expects each system’s clock to be synchronised with the clocks of the systems connected to it, and access to the clocks restricted to authorised people, so the times recorded across connected systems agree.