DRIVER recommendations – Experimental procedures

Experimental procedures encompass all activities carried out during an experiment, which may or may not involve the experimental model. They describe what was done, when and how often it was done, and which equipment and reagents were used. These procedures should be planned, optimised and tested before experiments begin, and then reported in sufficient detail to enable verification and reproducibility.

Establish and document all experimental procedures

Creating and maintaining detailed protocols helps ensure experiments can be reported with sufficient information to support reproducibility and remain current as methods evolve.

Protocols may be developed as part of an experiment or may follow pre-established methods published by others. Experimental procedures are most commonly sourced from the methods sections of published articles, with other sources including dedicated protocol repositories, manufacturers’ instructions for kits and equipment, and internal lab protocols. Any deviations or changes to established protocols, including procedures, equipment and materials, should be recorded in detail to support future reporting.

Information that should be considered during experimental design and recorded throughout the study include:

• Reagents: Source, preparation, quantity and concentrations.
• Equipment: Make, model and calibration.
• Models: Details of the model used (see Item 3: Experimental model).
• Timings: Duration, frequency and scheduling.
• Step-by-step protocols.
• Software: Versions, settings and bespoke code.

For further explanation of these components of experimental procedures see the interactive content (i. What makes up an experimental procedure?).

Experimental protocols can be recorded using protocol repositories, allowing procedures to be documented with step-by-step instructions in full detail. Various publicly available repositories dedicated to protocols are available:

• bio-protocol.
• protocols.io.
• STAR Protocols.

Ensure equipment and reagents are fit-for-purpose

All constituent parts of experimental procedures – including reagents, materials, and equipment – should be appropriately tested to ensure they can fulfil their intended roles within the procedure, whether as an intervention, a means of generating a signal or for capturing or measuring data. They should also be checked to ensure they are functioning as expected. Failure to do so may introduce error or bias (see Item 2: Risk of bias). These checks can be broadly grouped into the following categories:

• Quality control: Routine checks to ensure reagents, materials and equipment are performing consistently and as expected.
• Fit-for-purpose assessment: Specific testing to demonstrate that reagents, materials, methods or equipment are suitable for achieving a defined experimental outcome (e.g. demonstrating that an antibody specifically binds the protein of interest).
• Calibration: Checking and, where necessary, adjusting equipment measurements against known reference standards to ensure accuracy.

For expanded explanations on these checks, see interactive content (ii. Quality control, calibration and fit-for-purpose assessments).

Quality control should be an ongoing process embedded within experimental procedures and may be as simple as checking that reagents are within their expiry dates or stored correctly. Fit-for-purpose assessments should be undertaken whenever a reagent, material, method or piece of equipment is newly introduced or modified. The frequency and timing of calibration depend on the type of equipment and its use (e.g. frequency of use or the importance of the measurement to the experimental outcome) but should occur at appropriate intervals to ensure consistent and accurate results.

Reagents and kits should undergo both quality control and fit-for-purpose assessment to confirm that they are performing as intended. For biologics, new batches or lots should be tested to ensure they do not introduce unwanted variability into experimental results.

Equipment should be checked and calibrated prior to use in experiments. Failure to do so may introduce performance related or detection bias, depending on the role of the equipment within the experimental procedure.

Report experimental procedures in detail

Experimental procedures should be reported in sufficient detail to allow others to verify your results and replicate your experiments. Step-by-step instructions should be provided where needed, particularly for new or complex methodologies. 

When reporting experimental procedures, the citing of previous papers should not be used as a substitute for methodological detail in the manuscript. Key information required for replication is often missing from published articles. Full details should either be included in the manuscript or in a publicly available protocol which can then be cited.

Supplementary materials are often used as a place to provide expanded methodology sections or validation data. Although this is preferable to omitting the information altogether, supplementary files are not always assigned stable digital identifiers and are often difficult to access, search or reuse. Where possible, this information should be moved into the main manuscript or uploaded to a repository and cited within the manuscript.

Indicate what, when and how procedures were performed

What has been done, when and how often, should be reported, as well as the specific equipment and reagents used at each stage of the experiment. For complex procedures, diagrams, flow charts or other graphics may help to report the experimental design and procedures clearly.

Recording and reporting detailed information of how reagents were sourced, prepared and used is vital. Where available Research Resource Identifiers (RRIDs) should be used when reporting the reagents used. They are unique, stable digital identifiers allowing unambiguous identification of the reagents. 

Test articles are any reagent used for interventions or treatments during experiments. As the effects of the test article are usually what is being observed or measured in an experiment it is vital to provide full details of how these were sourced, prepared and used. Important details include:

• Unique identifiers.
• Source.
• Purity.
• Dosage.
• Concentration.
• Duration of use.

For chemical substances unique chemical identifiers from InChl can be used. 

For more information on test articles and what to include when reporting them see the interactive content (i. What makes up an experimental procedure?).

For biological reagents it is important to record and report lot numbers, as there may be variation between lots. Quality control checks should be performed on new lots to check for variation, particularly if different lots are used within the same experiment as they may introduce a source of bias (see Item 2: Risk of bias). 

Where specific equipment is required as part of the experimental procedures, details of how the equipment has been set up and calibrated is needed to allow others to replicate the same experimental set-up. Details of reference standards and versions of the hardware and any associated software should also be recorded. 

For experiments relying on robotics and/or imaging technology details of the instrumentation and software packages used, any custom scripts used for analysis and any post-acquisition data processing should be documented and reported.

Any software used, including statistical software, together with its version and details of how it was used, should be recorded. Versions of any packages or libraries used should also be documented. Custom scripts and code should be deposited in a suitable repository and assigned a DOI.