Are users requesting different units for the same specimen type?

[Donna_Housley]

The issue of how to standardize quantity units for Specimen Types has been an onging puzzle for me and recent discussions after OSCON23 has assured me that I’m not alone in my struggles. Different lab groups have strong opinions about which units should be used in the quantity field. For instance:

• Molecular Class: Volume (uL) vs. Yield (ug)
• Cells: Volume vs. Total cell count
• Tissues: gram vs. ‘bite’

If there is enough interest in this problem, perhaps the Krishagni Team would consider putting efforts into developing a standard system-wide solution. So please weigh in if this is important at your institute. If you don’t struggle with this issue, please share your solutions. Several of us would be happy to hear any ideas you have. I will include my solution in the comments section.

Thanks for any input!
-Donna

Donna Housley, PhD
Biospecimen Navigator
Clinical Translational Science Institute
University of Utah

Link to the last thread on this topic, which I imagine was addressed by allowing the system admin to set a system-wide unit in the properties feature for PVs.

Appropriate units for specimen types – molecular (eg. DNA/RNA)

Posting on behalf of @Peter_MacIsaac: One of our tissue banks has requested an internal change to OS default units for the molecular class of specimen types (eg. DNA/RNA) from micrograms ug to nanograms ng. Seeking input from other institutions about how you manage units for this type of micro specimen/aliquot The issue is that OS allows only a single default unit for each of the 4 major specimen type classes for mass and concentration. If we change then we have to identify all previous spec…

[Donna_Housley]

These are a few examples of how I’ve attempted to solve this for users at the University of Utah:

• Nucleic Acid: Since ‘yield’ is always a calculated number based upon volume and concentration, our solution has been to record volume in the quantity field, record concentration as that returned from nanodrop or qubits, and use a temporal query to display ug based on concentration and volume. A similar approach can be used for the Cell class (e.g. PBMCs).

• Tissues: some studies biopsy small tissue bites that are not weighed, they just record the number of bites. Others might weigh a tissue sample. I created two specimen types to represent the units requested (i.e. fresh tissue (qty = ‘gm’) and fresh tissue bite (qty = ‘bite’). Both of these solutions have drawbacks in terms of standardization of data when a researcher wants to find available tissue samples. They are technically both fresh tissue.List item

• Stool: within a single protocol we have a lab that receives solid stool samples (gm) and liquid samples returned in a vial for nucleic acid prep (mL). So I have one stool type under tissue and another under liquids.

All of these examples lead to my question, 'Can we have two standard fields, one for volume and one for quantity? Tissues would be quantity (number of bites/pieces) and the other would be ‘mass’(?). I don’t even know if this is possible from a coding standpoint, but from a biology standpoint, this is what we are used to using.

I look forward to hearing other solutions you might have!
-Donna

[Amritha_Menon1]

Hi Donna,
Thanks for raising this issue. We have the same problems as you.

(1) DNA/ RNA - We have one group who want to use the default OS units of quantity in micrograms and Concentration for DNA / RNA and another group who want to show Volume in uL and concentration for DNA/RNA and calculate quantity from it. We really do not want to create two different DNA types to deal with this.

Our solution would be for OpenSpecimen to have Volume, Quantity and Concentration as the standard fields. If the user enters two of these fields, the third should be calculated automatically. When aliquots or derivatives are created or samples are distributed, both quantity and volume should be automatically recalculated.

(2) PBMC – Some groups want to count the cells and measure it in millions of cells. Others want to keep it in ml.

How can we cater for two different units? Can we have this set up at the collection protocol level?

(3) Queries – We would like the specimen types units to be displayed in queries.

If this is done, then it does not matter if different protocols use different units for the same type.

Thanks,
Amritha

Amritha Menon
BRIDGE Technology Manager
Irving Institute for Clinical and Translational Research
Columbia University Medical Center
622 West 168th Street
New York, NY 10032
E: am4637@cumc.columbia.edu

[Specimen_Data_Fred_H]

This has also been a challenge for us at Fred Hutch. We have a workaround solution but I will admit it’s not the best. We add/manage specimens as needed by groups. When it comes to a conflict in measurements, we enter them twice.

For example, I have a few groups using Leukapheresis Product as a specimen type, but some measure in Cells and others in ml. Our solution was to enter it twice as Leukapheresis Product (Cells) and Leukapheresis Product (ml). It does mean that when doing any sort of metrics measurements, we have to ensure we count both specimens as one. But it does manage the issue of needing multiple measurements.

-Best,
Cleavon Joseph
Product Manager
Office of Translational Research | Research Administration
Fred Hutchinson Cancer Center
cjoseph@fredhutch.org

[Lee_Ann]

Hi all, we have this issue as well and tell users the measurements cannot be changed since it would affect all previous collected specimens. It would be nice if measurements could be adjusted by site or protocol. This would be for quantity and concentration. Our volume is capture in a form and the units are in the workflow. So when a lab requests unit changes for volume, all I have to do is download the json, update the unit measurement and reupload to the protocol.

Here’s a snippet below:
{
“name” : “specimen.extensionDetail.attrsMap.NT4”,
“caption” : “Volume (mL)”,
“type” : “text”,
“pattern” : “/^([0-9]+|[0-9]\.?[0-9]+[e]?[±]?[0-9])$/”,
“optional” : true,
“defaultValue” : “1.2”,
“showIf” : {
“op” : “OR”,
“rules” : [ {
“field” : “specimen.specimenClass”,
“op” : “==”,
“value” : “‘Molecular’”
}, {
“field” : “specimen.type”,
“op” : “==”,
“value” : “‘Fresh Cells’”

[Donna_Housley]

Thanks Lee Ann. I am faced with this exact issue again today (PBMCs). And yours is the only solution I can come up with to allow the lab to record a volume and a concentration. The only issue is that I’d like the system to calculate the cells from the volume * concentration. But as far as I know, you can’t set a standard field like ‘quantity’ as a calculated field. So the calculation to input cells still has to be done outside the system. I think the paid workflows app is the only workaround. We are hoping that if there is enough need for this, the Krishagni team might consider a system solution for this in the standard configuration. Thanks for providing the code snippet you used. I suppose I can just display the total cells in a query using a temporal query instead of recording it in quantity. Sadly that kind of invalidates all the useful features of the quantity field.

I think the main misunderstanding is that ‘Cells’ and ‘Yield’ (molecular: total ug) are both calculated fields in the industry. Neither is something that is a direct reading from any available technlogy. Both are based on concentration * volume.

Thanks for your feedback!