parts per quadrillion

[timbrisc]

Issue migrated from trac ticket # 5825

component: organization | priority: minor | keywords: ppq

2022-06-07 21:06:59: william.hess@fda.hhs.gov created the issue

---

UCUM Board needs to provide input. UCUM already has parts per thousand, parts per million, parts per billion, and parts per trillion, so it would be consistent to also add parts per quadrillion.

This new concept is needed to support the U.S. Food and Drug Administration Pharmaceutical Quality Chemistry Manufacturing and Control (PQ/CMC) initiative. Please see https://www.regulations.gov/document/FDA-2022-N-0297-0001, which in part, states:

"PQ/CMC is a term used to describe manufacturing and testing data of pharmaceutical products. PQ/CMC encompasses topics such as drug stability, quality specification, batch formula, and batch analysis, which are important aspects of drug development. PQ/CMC plays an integral part in the regulatory review process and life cycle management of pharmaceutical products. The development of a structured format for PQ/CMC data will enable consistency in the content and format of PQ/CMC data submitted, thus providing a harmonized language for submission content, allowing reviewers to query the data, and, in general, contributing to a more efficient and effective regulatory decision-making process by creating a standardized data dictionary.

The impetus for this standardization effort was the provisions from the 2012 Food and Drug Administration Safety and Innovation Act (Pub. L. 112-144), which authorized the Agency to require certain submissions to be submitted in a specified electronic format. PQ/CMC standardization supports FDA's regulatory needs in receiving structured and standardized data in pharmaceutical quality and includes two objectives: (1) To standardize the pharmaceutical quality data that is currently received by FDA in eCTD Module 3 from the sponsoring organizations, and (2) to use these structured elements and develop a FHIR data exchange solution."

[dr-shorthair]

Add [ppq] or [PPQ] to Table 3

[gschadow]

need reference to literature where this is actually used.

Quick google search makes PPM common, but PPQ is a gun. If there is very little use for it in the literature, it should possibly not be included. I request at least 3 articles full text where we can see how the real world deals with "PPQ". If it is only used as an abbreviation and 3 or 3 articles contain the definition as a concentration of whatever ng per liter or what it may be, then I vote to reject this. If it turns out that it is very common, by all means, include it.

[colin-e-hscic]

ppm/ppb/(and now ppq) are redundant as 10^-6, 10^-9 etc. suffice. Also "ppb" is ambiguous given the problem with "long" and "short" billions, so best to avoid the whole pattern.

[dr-shorthair]

Yes redundant, but widely used. ppb is very common in environmental chemistry.

Since we have multipliers to SI all round, in principle many customary units are redundant. They could be expressed using integers above and below a solidus, but we don't do that.

I don't think the goal of UCUM is parsimony. It is uniqueness and convertibility.

[colin-e-hscic]

I'll put forward an alternative viewpoint. UCUM already includes a lot of units that are historical, obscure or otherwise not things you expect people to be using in regular technical/scientific environments.

In theory at least the more "weird stuff" is included in UCUM without an option to select a modern rational unit subset, the more chance of typos or other errors happening to pass validation because they collide with one of these pieces of esoterica. I accept the square brackets [unit] disambiguation syntax helps avoid that, but IMHO that is beginning to be applied in ways that perhaps aren't ideal.

Perhaps there are two needs competing here -

One is to be able to "say anything people might want to say about units" in UCUM, which argues for expanding the standard indefinitely.

The other is consider UCUM as a domain-specific programming language. I think the experience with languages design is that-

a) Languages tend to grow, accreting features until they become unweildy and users head off looking for a cleaner simpler alternative b) 90%+ of the capability comes from 20% of the feature set c) 90%+ of the bugs come from the obscure 20% of the language most users don't really understand

[colin-e-hscic]

As a a further comment, UCUM is focused on machine to machine communication of units in an unambiguous and computer-processable form. For that purpose the smallest and easiest to understand/memorise/use language is the best.

Users in a particaular field might like to use a specialist set of units, almost as part of a "dialect", but UCUM isn't designed for users, so if there's a good equivalent for a requested unit available as an expression using the existing UCUM atoms, the commuinication requirement is met. Users might want to see "ppm" but the machine is quite happy with "10^-6"

It would be great if UCUM included both a fully computable machine syntax AND a fully computable human-readable syntax (plus a bidirectional transform between them of course), but for the moment it doesn't so we have to find other approaches. We are currently planning to communicate both a UCUM expression and a separate (manually edited) human readable display unit for measured values in our environment, but that might not work for you.

[gschadow]

May I ask who Colin-e-Hscic is? I get pseudonyms in public forums, but here it would be nice if we knew with whom we're talking. Maybe you could sign you name under messages?

I appreciate push-back against weird units. But we have to keep an eye on reasonable continuity. There is space to do complete re-thinking from time to time, but such complete re-thought cannot really change the course on a particular decision which is supported by precedence.

I have been in committees a lot of my life, and I think there is an overlooked demand for reasonable method of decision making.

So, the issue is here ppq. I see no follow-up establishing how common it is. The reason ppm and ppb have been included is because they are extremely common in any sort of toxicology laboratories, environmental, pharma, or otherwise.

UCUM is for both, computer to computer and human to computer interfacing. Otherwise we would not have any symbols or grammar, we could do everything with number and exponent vector. So for that human interface, we do care about what people actually use. But we do not agree that every abbreviation ever used by anyone somewhere should become a UCUM symbol. There has to be a certain level of support. Can that be quantitatively objectified? In my above comment I have put the onus on the requester to produce 3 pieces of evidence that shows the use of such unit symbol as a well known term (This is analogous to the Patent and Trademark Office requiring specimens of use of the trademark in order to register it.). To be a well known term, there must be evidence that it is understood in a certain sizable subgroup other than just one running abbreviation defined in an article. Best examples is "kg". A scientific article can easily write "20 kg" without ever defining "kg" as an abbreviation. People don't write "20 kilogram (kg)". And with "ppm" and "ppb" I say that that also goes like that. Nobody writes "we measured lead contamination in parts per billion (ppb)", no, they use ppb. Now, for ppq that begins to be in doubt. Either it can be established by evidence, or not.

And that is how this decision should be made.

The other, wider questions if those are "redundant" units are secondary. Liter is also redundant to dm3, so redundancy isn't really an argument against a common unit. Also, a general fear of ambiguity is not, after all, we bracket such units, writing "[ppm]" not just "ppm". Since the beginning of UCUM we had paid attention to these ambiguities, which was the original reason to invent this bracket notation. So this is deeply in UCUM's DNA, we're not going to make that error now.

[dr-shorthair]

May I ask who Colin-e-Hscic is?

The best place to record some identifying information is in each user's GitHub profile. @gschadow and @colin-e-hscic perhaps you could both give a little more there. (Yes, all the UCUM old hands know Guenther, but newcomers may not.)

[colin-e-hscic]

May I ask who Colin-e-Hscic is? I get pseudonyms in public forums, but here it would be nice if we knew with whom we're talking. Maybe you could sign you name under messages?

Fair question. That id was freshly created recently to link to our new organisational single signon setup on Github, so it is lacking in both Bio information and activity history. I can do something about the first part of that. I'm not a significant author on Github, but my other login ids do at least have a few more years of history.

[gschadow]

May I ask who Colin-e-Hscic is? I get pseudonyms in public forums, but here it would be nice if we knew with whom we're talking. Maybe you could sign you name under messages?

Fair question. That id was freshly created recently to link to our new organisational single signon setup on Github, so it is lacking in both Bio information and activity history. I can do something about the first part of that. I'm not a significant author on Github, but my other login ids do at least have a few more years of history.

The funny thing is, after this answer I still don't know.

[colin-e-hscic]

UCUM is for both, computer to computer and human to computer interfacing.

On that point I was guided at least in part by the UCUM documentation. From the "What is UCUM" section of the ucum.org web site-

The focus (of UCUM) is on electronic communication, as opposed to communication between humans.

And also by the challenge of developing a standard UCUM code list (yes another one i'm afraid...) for UK Pathology. For that work we have decided we need to tabulate both a preferred UCUM code and a corresponding display format representation.

[colin-e-hscic]

The funny thing is, after this answer I still don't know.

Because I haven't had a chance to update my profile yet, I was typing another answer above!

I you are saying you'd like a direct response here, I am a Business Analyst at NHS Digital in the UK. I have been working on our plans to implement UCUM coding of UoM as part of our rollout of new messaging standards based on FHIR and SNOMED for UK Pathology.

[colin-e-hscic]

If this is intended to be a closed group I apologise for crashing the party. I have been working on this for a while and seeing some of the subjects raised here triggered some immediate thoughts.

[gschadow]

If this is intended to be a closed group I apologise for crashing the party. I have been working on this for a while and seeing some of the subjects raised here triggered some immediate thoughts.

No, no, it's open. You can chime in. As you did.

For that work we have decided we need to tabulate both a preferred UCUM code and a corresponding display format representation.

So you see, the issues are not completely separable. Again, if it was only about computers, a simple number x exponent vector representation would be sufficient. We have the symbols because that is how humans think and write, and we support that.

[colin-e-hscic]

UCUM is for both, computer to computer and human to computer interfacing. Otherwise we would not have any symbols or grammar, we could do everything with number and exponent vector.

That sounds more like an argument for a high level vs. low level language. The number and exponent vector (plus the function hook for special units) is the machine code, and UCUM is the high level syntax. A high level language is clearly a boon for developers, a lot more people use Javascript now than assembler or even C, but it's not a strong argument for putting Javascript or Perl in front of users when what they want to know is what the program behaviour is (its meaning).

Clearly UCUM syntax is far closer to a "human readable" form than most languages, at least for simple cases. However that can be a two-edged sword when you run into an example that looks simple but has a hidden bear trap.

[SRKid]

Citations where the federal government and private sector have used parts per quadrillion (ppq or PPQ):

DEPARTMENT OF HEALTH AND HUMAN SERVICES, Food and Drug Administration, 21 CFR Part 176 [Docket No. 96F–0401] Indirect Food Additives: Paper and Paperboard Components ACTION: Final rule. Federal Register / Vol. 63, No. 219 / Friday, November 13, 1998 / Rules and Regulations https://www.govinfo.gov/content/pkg/FR-1998-11-13/pdf/98-30296.pdf

Detection of Melamine from Food in Parts per Quadrillion Level Using Functionalized Graphene Oxide-Gold Nanoparticle Hybrid SERS Platform https://www.researchgate.net/publication/285393050_Detection_of_Melamine_from_Food_in_Parts_per_Quadrillion_Level_Using_Functionalized_Graphene_Oxide-Gold_Nanoparticle_Hybrid_SERS_Platform

Dioxins in commercial United States baby food J Toxicol Environ Health A, 2002 Dec 13;65(23):1937-43. doi: 10.1080/00984100290071450 https://pubmed.ncbi.nlm.nih.gov/12490040/

U. S. Environmental Protection Agency, Ground Water Program, 8P-W-GW https://www.epa.gov/sites/default/files/documents/MSU-ERI_Draft_SOB.pdf

U. S. Environmental Protection Agency, Interim Drinking Water Health Advisory: Perfluorooctanoic Acid (PFOA) CASRN 335-67-1 https://www.epa.gov/system/files/documents/2022-06/interim-pfoa-2022.pdf

California announces bold public health goals for PFOA and PFOS in drinking water https://www.ewg.org/news-insights/news-release/2021/07/california-announces-bold-public-health-goals-pfoa-and-pfos

Part per Quadrillion Measurement of Explosives and Other Chemicals, SilcoTek https://www.silcotek.com/blog/part-per-quadrillion-measurement-of-explosives