IT systems · Practical guide

Chromeleon: tracing changes to sequences

A sequence connects injections, samples, methods and processing. Changing these relationships can change the interpretation of an investigation even when the underlying measurement data remain intact.

Table of contents

At a glance

How do you assess changes to a Chromeleon sequence?

Assess whether the Chromeleon sequence change affects identity, order, processing or completeness of the analysis. Include added or aborted injections, changed sample assignments and the method version. The reviewer needs to reconstruct relevant events. The final sequence view alone may not show the full sequence of actions.

Scope and Applicability

A 2018 Thermo Fisher white paper describes an Injection Audit Trail for each injection, including permitted manual interventions during raw data acquisition. A change to a running sequence is attributed to the person making it. Test the events and views available for your version and configuration in the actual workflow. This article does not prohibit permitted laboratory changes.

Sources: [1] [2]

The change must apply to the correct sample and at the correct time.

The allowable modifications arise from the laboratory procedure. Therefore, document the rationale, decision, and handling of all relevant data. The potential consequences of an addition, renaming, or retrospective processing are addressed in the following section.

Review the completeness of the review using a deliberately altered test workflow. What events are visible in the data log or in other logs? What views does the reviewer require? A large volume of logs is only useful if critical changes are actually identified and assessed.

Sources: [1] [2]

Classify sequence changes according to timing and effect

A sequence combines samples, injections, methods and processing. A change can affect preparation, on-going recording or subsequent evaluation. Therefore, describe not only that something has been changed, but when, by whom and with what effect on the actual process.

Capture the deployed Chromeleon version, project structure, roles, and audit trail context. Manufacturer documentation outlines various forms of evidence; it does not automatically validate your specific configuration. Review which changes appear in each protocol and determine whether a reviewer can achieve the complete required context.

Distinguish between expected modifications and unexplained changes. An additional injection may have a legitimate technical justification. A name change may mask a misassigned sample. Both actions should be evaluable within the intended procedure. A blanket prohibition against any change is no more effective than a simple reference to existing audit trails.

Sources: [1] [2]

Prepare a change matrix for critical sequence components

Name relevant fields and actions. This may include sample identification, position, injection, method reference, and result usage. For each element, define who is authorized to make changes and the associated documentation or evaluations required. The list shall be aligned with your laboratory process, not with a universal checklist of all software functions.

Consider changes that occur after data entry has begun. A later-visible sequence status may differ from the actual executed status. Therefore, evidence should explain which injection was performed under which context. A current table view alone may be insufficient for this purpose.

Internal Work Aid: Develop a change matrix for critical sequence components
Type of changeProcess questionIn-house verification
Sample identification Does the injection still belong to the correct sample? Reconstruct original and amended references
Methodological basis Which method was actually used? Separate consideration of data entry and data analysis status
Additional injection Why was it repeated? Understand the rationale and outcome of usage
Sequence Does the procedure's significance change? Compare the version executed with the plan
Hiding or selecting data What data are included in the report? Verify the completeness of the required review scope.

Sources: [1] [2]

Example: a repeat injection following an anomalous result

The sequence includes sample P-31. Following an atypical result, an additional injection is performed. The internal requirement stipulates that the original injection, reason for repetition, repeat injection, and the resulting data must remain traceable. The analytical validity of the repeat injection is assessed in accordance with the applicable laboratory procedure; this document does not provide a universal analytical decision.

Prepare the sequence with a known starting condition. Add the repetition via the intended pathway and document user, timestamp, and relevant change. Then verify whether another authorized reviewer can reconstruct the process without additional verbal explanation. The test should not only demonstrate the technical initiation of the new injection.

Include an unauthorized change to the sample identification and a report containing only the final result. Verify whether necessary context remains visible or can be purposefully retrieved. A professionally justified report may select data; the underlying data basis must not become incomprehensible or incomplete.

Own work aid – Example: a repeat injection following an atypical result
Test caseActionExpected evidence
Starting sequence Record the specified version Probe, injection and method clearly defined
Justified addition Create a repeat with intended role assignment Change and rationale traceable
Unauthorised change Attempting to change critical attribute The selected access control works.
Historical Investigation Reconstruct the original workflow The original state and later changes remain distinguishable
Results report Select and context review The result used is justifiable and traceable.
Review Engage an independent qualified reviewer Required information actually accessible

Sources: [1] [2]

Verify audit trail filter completeness

A filter facilitates review but may obscure relevant events. Derive the selection criteria from critical changes. Verify with known test events whether the required actions appear in the expected report. An empty report is only meaningful when data flow and selection rules function correctly.

Consider the time window and object assignment. An event may fall outside a narrowly defined window or be assigned to a different object context. The reviewer should understand the limitations of the report. A clean presentation is not a substitute for known coverage.

Gain the ability to return to detailed context in response to anomalies. A summary may serve as a suitable entry point; a detailed investigation of a specific event may require additional information. Access rights, available data, and retention requirements must support this approach.

Sources: [1] [2]

Evaluate roles and subsequent processing together

Capture who is authorized to modify sequences, process data, and select results. Verify actual user combinations and privileged access rights. A role with individually defined restrictions can be extended with additional authorizations. The test should reflect the actual user base in use.

Distinguish between sequence change and processing change. An unaltered injection sequence may subsequently be re-evaluated differently. Conversely, an additional injection can be performed without modifying the evaluation method. Both approaches may influence future conclusions and require appropriate linkage in the review.

An electronic approval or signature should refer to a specifically identified content item, if used within a procedure. The name of a report alone does not fully describe the content involved. Verify which data and versions actually underpin the decision, and ensure this remains traceable for future review.

Sources: [1] [2]

Changes to the system and handover to operations

For new releases, roles, or report templates, assess whether the traceability of critical sequence changes is affected. Maintain known reference events for regression testing. A test focusing solely on device communication does not confirm the integrity of the audit trail evaluation.

Submit the change matrix, filter limits, and testing pathway to reviewers and system owners. A practical exercise using an augmented injection and a modified identifier demonstrates whether the context can be identified and assessed. Participation alone in a general software presentation only provides limited evidence of this specific capability.

Obtain required context even during archiving or migration. A chromatographic numerical value without associated capture and processing may lose its analytical significance. Therefore, subsequent access should appropriately retain the context necessary for your process.

Sources: [1] [2]

Draft: Track sequence changes from original plan to review

Transfer the case described above to the search template. A legitimate reviewer should be able to understand the initial plan, actual injections, changes and use of results from the information provided. Additional oral explanations of the analyst may indicate a gap in report, access or procedure.

Is information missing? Investigate the root cause: was it not captured, not retained, or merely inaccessible? A broader report may address the latter scenario, but does not restore a history that was never recorded. Document the actual, verified coverage.

Sample filling example – internal work aid
Search fieldExample entryExpected finding
Plan Original sequence with sample P-31. The planned starting state is known.
Data capture First injection and method reference. Actual generated data assigned.
Change Additional duplicate injection. Person, time, and relevant reason clearly traceable.
Processing Identified analysis versions. Sequence and processing changes distinguished.
Selection Result used in the report. Technical justification and context available.
Filter Known event appears in the review report. Selection does not conceal critical process.

Sources: [1] [2]

Next steps

  • Define allowable changes and required justifications.
  • Assign affected data objects and protocols.
  • Perform targeted modifications, interruptions, and repetitions.
  • Check the completeness and clarity of the review.
Further reading:

Project experience and collaboration

Reconstruct the sequence change from the reviewer’s perspective

With the laboratory, we prepare a test sequence with a known starting state and a justified change. Another authorised reviewer uses the available views to reconstruct the injection, change, user and result used. A verified filtered report may support that review. If events or important relationships are missing, we improve log access, reporting or the procedure as appropriate. Widening a report filter cannot recover an event that was never recorded.

Frequently asked questions about Sequence changes in Chromeleon

Is each additional injection considered a deviation?

Not automatically. The reason and the approved procedure determine the action. Unplanned or result-relevant changes may require a separate evaluation.

Sources: [1] [2]

Is the latest sequence report sufficient?

This depends on its content. If relevant changes, discontinued trials, or context are missing, additional records are required for review.

Sources: [1] [2]

Must the reviewer be an administrator?

Not inherently. He requires appropriate read rights and views for his task. Unnecessary change rights should not be granted solely for review purposes.

Sources: [1] [2]

Are sequence alterations fundamentally unacceptable?

No. Changes may be technically justified and permitted under the intended procedure. Key considerations include authorization, timing, rationale, and impact. Verify that both the original and modified procedures remain understandable and that the resulting outcome is properly supported. The presence of an audit trail function alone does not substitute for this technical evaluation.

When does a filtered report suffice for audit trail review?

When professional selection, coverage, and boundaries are justified and validated. Generate known critical changes through relevant roles and pathways, and compare the expected events with the report. Include time windows and objects. Upon detecting anomalies, the required level of detail must remain accessible. An empty report neither confirms completeness nor verifies the absence of relevant changes.

Sources: [1] [2]

Is the current sequence view sufficient as historical evidence?

It may reveal important information, but does not automatically reflect the originally planned or actually captured status. Review required history and changes. The reviewer should be able to identify which injection was generated under which context and which result was subsequently used.

Sources and review date

The sources provide the technical foundation. Tables, checklists, and examples are derived methodological tools, not verbatim regulatory requirements. Manufacturer information must be referenced to the specific version and configuration in use.

  1. European Commission: EU GMP Annex 11, Computerised Systems (2011)Current version as per EudraLex inventory as of 14.10.2026. Application, risks, operation, and data lifecycle; specific testing proposals are derived independently.
  2. Thermo Fisher Scientific: Data integrity – audit trails with ease of review, White Paper 72664 (2018)White Paper 72664, © 2018; p. 9: Injection Audit Trail and modifications to ongoing sequences; p. 12: Sequence context and audit trails. Manufacturer description – to verify against the version and configuration state in use; no blanket compliance certification.

Sources checked: .

Daniel Herrmann, Daniel Herrmann Consulting

Daniel Herrmann Consulting

Check sequence changes and review coverage

We support your laboratory in reviewing critical sequence modifications, available protocols, and required reviewer views at the deployed site.

  • Computer system validation →

    Bring your GxP systems into validated operation, from laboratory software, MES and eQMS to ERP. We define the scope, plan risk-based tests using GAMP 5 and hand the evidence to QA for release.

Send a project enquiry Customer testimonials →