New term proposal : isotopolog_incu_time

[mslarae13]

New term details For us to assess a new term request we require the following details:

Term name - isotopologue incubation time
Structured comment name - isotopolog_incu_time
Definition - Total time of incubation after isotopologue addition in hours
Expected value - value
Value syntax - {float}{unit} 
Example - 72 hours
Preferred unit - hours

Additional context Add any other context about the new term here.

[simpso91]

Term name - isotopologue incubation time Structured comment name - isotopolog_incu_time Definition - Total time of incubation after isotopologue addition in hours Expected value - value Value syntax - {float}{unit} ((Would this be better as a timestamp or duration?)) Example - 72 hours Preferred unit - hours

[mslarae13]

How common is it to capture duration? I think duration can be inferred from am time stamp..collection date and time - incubation time... but! most people rarely record collection time.. so duration would be hard / unlikely to be true.

What's better / more useful for the community? Time and date it was added? Or how long the isotope exposure went on?

@simpso91

[Woolly-at-EBI]

(interesting, record collection date/time will vary by community: in the marine community, record collection date/time is much more important and thus often collected. e.g. given the GPS coordinates+date+time oceanographic like light levels (even night or day!) and low/high tide can be fairly reliably derived. I guess this matters far less for your community, but would still expect the date at least to be useful in analysis.)

In clinical(pharma) where I previously worked, both sampling time and "treatment" duration were widely collected. sometimes both and sometimes only one was present. I am pleased to see are defining the units as hours, in the clinical trial/observational world there was a less consistency in practice. )

[simpso91]

Most SIP papers report how long the incubation went on in hours/days/week. Many studies also specify date the environmental sample was collected, but few specify what date the incubation started or ended. I think incubation time (instead of date) would be most useful, especially if people tend to leave out collection time. For SIP, isotope crossfeeding (which increases with incubation time) can completely change results, so incubation time might be more important for interpretation.

[mslarae13]

@turbomam Can you comment on the format we should collect incubation duration?

HH:MM:SS. ... DD:HH:MM:SS. ? So 5:17

or

5 hours 17 minutes

or ????

[only1chunts]

Most terms say they expect times in Iso 8601 standard format, how many values are actually given in the correct way is a different matter! Sfor your example of 5hours17mins, its a time value so should be preceded by the T indicator: T5:17:00

On Fri, 16 Jun 2023, 22:14 Montana, @.***> wrote:

@turbomam https://github.com/turbomam Can you comment on the format we should collect incubation duration?

HH:MM:SS. ... DD:HH:MM:SS. ? So 5:17

or

5 hours 17 minutes

or ????

— Reply to this email directly, view it on GitHub https://github.com/GenomicsStandardsConsortium/mixs/issues/588#issuecomment-1595325968, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABOB5GO7Q2J6D6IGKHCMZBLXLTECNANCNFSM6AAAAAAYYYL6LA . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[turbomam]

Great question. Can we discuss this today?

Ideally, temporal data wouldn't just be saved as ISO 8601 strings, but rather as ISO 8601 objects, which is possible with most of the data serialization supported by LinkML. TSV/CSV is probably weakest in this area.

[turbomam]

The best way for stakeholders of all kinds to agree on a formal specification like this is to provide valid and invalid examples within a LinkML data file.

[mslarae13]

@turbomam looks like we forgot to discuss this! The next TWG meeting isn't until Aug 1st. Do we want to discuss then?

@ramonawalls FYI

[ramonawalls]

I'm going to be in Thailand on Aug. 1, so I think I am supposed to schedule another TWG call before then. Let me go work on that now.

[mslarae13]

https://www.digi.com/resources/documentation/digidocs//90001488-13/reference/r_iso_8601_duration_format.htm

(regex for validation '^P(([0-9]+Y)?([0-9]+M)?([0-9]+W)?([0-9]+D)?(T([0-9]+H)?([0-9]+M)?([0-9]+(.?[0-9]+)?S)?))?$' )

regexr.com/7hhbj you can test valid input here

For example, PT12H for 12 hours (P= period) 4 days = P4DT 1 week = P1WT

OR

Require in hours

12 or 4 days = 96

@simpso91 Preference is for the ISO standard. Thoughts?

[simpso91]

@mslarae13 Roli and I discussed leaving it in hours, because it is easier to use for users who aren't familiar with the ISO standard, and it is more intuitive (in lieu of parsing a whole string of numbers/letters).

We are open to pushback if you think users will list days instead of hours or if it makes the MIxS packages more consistent!

[mslarae13]

Discussed during the MISIP pub, this term is similar to https://genomicsstandardsconsortium.github.io/mixs/0001237/

We can either re-use this term, cleaning up the description or adding slot usage.