Why comprehensive data-package records matter
A complete LC-MS data package is more than a collection of spectra: it is the combined record that links acquisition context, raw electronic files, processing actions, and documented reviewer rationale. Clear records enable reproducibility, support troubleshooting, and preserve the rationale behind decisions made during analysis. Adopting systematic practices reduces ambiguity when reconstructing a package, especially in multi-operator or multi-instrument workflows.
Capture acquisition context at the time of data generation
Record the instrument configuration, column and solvent identifiers, lot or batch references for critical reagents, serial numbers, and timestamped run metadata. Where possible, retain machine-generated metadata and pair it with operator-entered notes describing deviations or notable observations. Embedding acquisition details in metadata files and maintaining a contemporaneous acquisition log ensures that raw spectral files can be tied back to the experimental conditions that produced them. See guidance such as the FDA guidance on data integrity and national frameworks for approaches to trustworthy recordkeeping.
Record processing history and rationale for transform steps
Processing LC-MS outputs typically involves several software stages—peak detection, integration parameter choices, smoothing, baseline handling, and export formats. For each change, log the timestamp, operator identity, software and version, and the parameter values used. Retain original unedited outputs alongside any reprocessed versions, and document the rationale for edits rather than relying on implicit assumptions. For example, note that a reviewer elected to reprocess a run after observing a transient baseline fluctuation, and cite the specific corrective parameters applied. This level of detail makes it possible to reproduce derived summaries and to understand divergences between successive outputs.
Document review decisions and maintain an audit trail
Review records should explicitly capture who reviewed the data, when the review occurred, a concise summary of findings, and the decision logic. Avoid shorthand; use plain, reproducible statements that list the evidence considered and any conditional criteria for revisiting a conclusion. Maintain versioned report files and ensure links exist from review notes to both raw and processed files. When signatures or approvals are recorded electronically, confirm that these are tied to system accounts and that system audit logs are preserved in accordance with applicable principles such as those described by the OECD for GLP data integrity.
Practical recordkeeping tips for reproducible LC-MS packages
Establish consistent file-naming conventions that include date, instrument identifier, sample ID, and processing status. Store metadata in machine-readable formats (for example, JSON or XML) alongside raw files, and preserve folder structures and permissions in backups. When exporting summary reports, include direct references—file paths or unique IDs—to the originating raw and processed files and to any scripts or parameter files used. For collaborative projects, define a minimal mandatory metadata set to capture at acquisition, processing, and review; consistency across contributors reduces reconstruction effort later. Additional principles and examples for maintaining trustworthy records are available from national guidance such as the UK GxP data integrity guidance and from international summaries like the OECD GLP data integrity guidance.
Not for human consumption. For laboratory research use only.