Sample and Inventory Management
TL;DR
Sample tracking means recording what each sample is, where it sits in storage, what state it is in and what has happened to it, from creation or receipt until it is used up or destroyed. Done well, the record takes you to the right box position and back through every parent sample to the original material.
Sample tracking is the practice of recording what a sample is, where it is stored, what state it is in, and what has happened to it, from the moment it is created or received until it is consumed or destroyed. Each sample gets a unique ID and a storage position, plus a status and a history, and its record links to the experiments that used it and to the samples it came from or gave rise to.
It started in LIMS software for QC and contract testing labs, where thousands of samples go through the same workflow every week and every one has to be accounted for. Research labs came to it later, usually after someone spent an afternoon in a -80°C freezer hunting for a cell line that had been in a colleague's box all along.
The term overlaps with lab inventory management and chain of custody. Inventory management covers everything a lab stocks, purchased reagents and consumables included. Sample tracking is about the material the lab makes or receives for testing (tissue, plasma, DNA preps, cell pellets) and how those samples relate to each other. Chain of custody is narrower again: a signed record of each handoff between people.
As far as the records made at the bench allow. Say a postdoc isolates RNA from twelve mouse livers. Each prep gets an ID that is never reused, the metadata that describes it (source animal, species, concentration, extraction date), a freezer, rack, box and slot, and a status such as available, reserved, quarantined or gone. Every action after that goes into its history with a user and a timestamp.
Lineage is the hard part. The RNA becomes cDNA, the cDNA is split into aliquots for qPCR plates, and one aliquot goes to a collaborator for sequencing. Each child has to point back to its parent, so when one qPCR result looks off, someone can follow it through the cDNA to the liver and the animal it came from.
Spreadsheets cope with the first level of that and struggle by the third. Two people edit the file in the same week, or a tab gets sorted without the ID column, and the parent links are gone.
Then there is the link to experiments. A record that lists which notebook entries used a sample, with dates and amounts, answers the questions a PI asks eight months later: which batch went into Figure 3, and is there any left?
Sample tracking is a regulatory requirement for studies conducted under GLP. 21 CFR 58.130(c) requires specimens to be identified by test system, study, nature, and date of collection, and 21 CFR 58.107 requires that proper identification of test and control articles is maintained throughout distribution, with the receipt and distribution of each batch documented. The OECD Principles of Good Laboratory Practice set equivalent expectations outside the United States.
For labs that aren't running GLP studies, sample tracking is good practice, and it shows up in ordinary weeks. When a postdoc leaves, the next person inherits four freezer boxes with records that say what each tube holds and how it was made. And when a reagent lot is recalled or a cell line fails a mycoplasma test, those records are how the lab finds every downstream sample and experiment that touched it.
The one holding the samples you could least afford to lose: irreplaceable patient or animal material, master cell banks, reference standards, and anything shared with collaborators. Before entering a single sample, settle the ID format (a project prefix and a running number is enough) and the short list of fields every sample of each type must have. A handful of required fields holds up better at the bench than a twenty-field form.
Set up storage before samples. Build each freezer down to rack and box level, then work through one freezer at a time and give every box a label and a record. Printed barcodes save reading handwritten codes off frosted tubes. Record parent and child links from day one too, because adding lineage to two years of existing samples is far more work than capturing it at the moment of aliquoting. For biological material in particular, see Optimizing Biospecimen Inventory Management in Research Labs.
In IGOR, sample tracking sits in the Inventory module. Sample IDs are assigned automatically from the database prefix, the two-digit year and a running count, and storage is a nested Storage Locations hierarchy, with Box Map View showing the grid of each box. The relationship viewer shows parent and child samples as an interactive family tree, and Sample History records every action on a sample with the user and a timestamp.
A sample tracking system should record a unique sample ID, descriptive metadata, the storage location down to the box position, the current status, and a timestamped history of every action with the user who took it. Parent and child relationships for aliquots and derived samples belong in the record too, along with links to the experiments that used the sample.
Yes, up to a point. Excel works for a few hundred samples with one person maintaining the file.
It struggles with parent and child relationships and with audit history, because a spreadsheet doesn't record who changed a cell or what was in it before. Most labs move to a LIMS, or a combined ELN and LIMS, once samples are shared across people or projects.
Give the aliquot its own unique ID and record the parent's ID in the aliquot's record, ideally at the moment you aliquot. Reconstructing it later from tube labels is slow work. In IGOR, parent and child links appear in an interactive family-tree viewer, and logging usage on a quantity-tracked item to create a child sample builds the link automatically.
Sample tracking follows the material a lab generates or receives for testing through its lineage and use, while lab inventory management covers everything the lab stocks, purchased reagents and consumables included. Many platforms handle both in one module. The distinction matters mostly for lineage: a bottle of Tris rarely has a parent, and a plasma aliquot always does.