Skip to main content

Biomedical Engineering: Good Lab Practices

General guide to resources for biomedical engineering.

Why should you care?

Could another person find, interpret, and use your data?

  • Someone in your lab?
  • Someone in your field?

Do any of these apply to you?

  • larger-scale and multidisciplinary projects
  • teams of geographically separated scientists
  • you agree that scientific journals are no longer capable of supplying all the information needed to ensure reproducibility

Good Research Records

Research records are important for:

  • Research planning and management
  • Results replication
  • Documenting collaborations
  • Publishing and peer review
  • Complying with governmental and institutional rules and regulations 

Good research records should include:

  • research planning and protocol descriptions
  • data manipulations and analysis procedures
  • personal and group interpretations of the results
  • important communications and group decisions among collaborators 

Good research records: logs, notebooks, correspondence, videos, computer databases, audio or digital records, research data, etc.

Lab Notebook (LN) Recommendations

Laboratory notebooks are important for:

  • Data management
  • Support intellectual property rights
  • Accusations of research misconduct 

Ideal Lab Notebook should include:

  • Experimental metadata
    • Equipment used, set-up, etc.
    • Drawings/photos of set-up
  • Raw data
  • Data analysis
    • Analysis tools used
    • Graphs and figures
  • Cross-references and citations
    • Journal article, previous experiment in notebook, etc.
  • Research ideas
  • E-mails, letters, other discussions about research
  • Table of contents and/or index

Proper LN practices:

  • Bound book with pre-numbered pages
  • Contains a table of contents
  • All entries should be dated and initialed
  • Use permanent ink
  • Write legibly without abbreviations 
  • No blank spaces on any of the pages (if a blank space is unavoidable, it should have a line drawn through that section of the page)
  • Errors should have a single line drawn through it and be initialed by the researcher
  • Any data inserts must be taped, or preferably glued and researcher should sign and date the insert 
  • Must have corroboration (every page  should have spaces for both the researcher’s signature and a witness’ signature)
  • Proper storage of the active and completed notebooks 
  • Each experiment should be given a title and identification number
  • List all the experiments in the table of contents
  • Experiments should be ordered chronologically
  • A rationale should be included explaining why the experiments are being conducted
  • All essential elements used in the experiment should be listed
  • Use complete descriptions and accurately describe the methods used to carry out the experiments.
  • Abbreviations used that are not common knowledge need to be defined.
  • Experiment evaluation and conclusions 

Apply good practice principles in your lab

  • Plan for data management and sharing
  • Consider ethical and legal issues
  • Document your data with sufficient metadata
  • Use proper file formats to ensure long-term access
  • Establish files naming conventions
  • Consider data integration
  • Secure and backup your data
  • Archive your data
  • Everybody in the lab knows the rules and apply them consistently
  • Know funder, publisher, and university requirements and plan accordingly
  • Consider the benefits that following through with data management planning could mean to you, your advisor, your colleagues, and scientific world


  • File formats to ensure long-term access:
    • PDF/A, not Word
    • ASCII, not Excel
    • MPEG-4, not Quicktime
    • TIFF or JPEG2000, not GIF or JPG
    • XML or RDF, not RDBMS
  • Files organization strategy:
    • Directory Structure Naming Conventions
  • File Naming Conventions
    • Label files consistently
    • Descriptive but short name
    • Avoid “ / \ : * ? ‘ < > [ ] & $ characters
    • Use underscores not spaces
    • Date files and do it consistently (YYYY-MM-DD)
    • For analyzed data, use version numbers
  • Backup your Data
    • Make 3 copies (e.g. original + external/local + external/remote)
    • Have them geographically distributed (local vs. remote depends on recovery time needed
    • Automated whenever possible
    • Departmental / university servers
    • Lots of Copies Keep Stuff Safe (LOCKSS)