search

AAFC-BICoE / dina-planning

AAFC-DINA planning repository
3 stars 2 forks source link

Demo Meeting April 28 - Material Sample Naming Discussions Continued #327

Open ssbilkhu opened 2 years ago

ssbilkhu commented 2 years ago

Presentation from Tara & Cladia of some naming situations

Presentations are contained in #326

ssbilkhu commented 2 years ago

Discussion:

Samples live on the sample "Table" as Material Samples

Delay the naming standard until more search functionality is available.

Requirement for Search:

ssbilkhu commented 2 years ago

Comments from @ron-reade :

  1. For the first slide I prefer the alphabetical approach to naming children (and my reasoning is a bit lame), as you can see from my diagram, it is not uncommon for generations of the same material to get out of sync, so in the example provided, generation 4 for lineage “A” is at “K” and L where as for lineage “C” it is only at “Q” and “R”. If we had opted to use numbers for lineage, then “A” would be at 11 and 12 but lineage “C” would be at 17 and 18. This would make me think by looking at the names that 17 and 18 were somehow newer than 11 and 12, but this is not the case. I realize that for alphabetical there is a similar order, but for whatever reason my brain doesn’t make that same assumption.
  2. I have used dashes, but I recall some discussion regarding dashes versus underscores (I am not invested in this).
  3. For subsamples I would like to see an alphanumeric structure so that I can give my subsamples a prefix that identifies the type of material that the sample is (R=RNA, P=PCR, S=Sequence ETC) see slide#2
  4. You can see that within a group subsamples are tracked numerically, but the naming allows me to directly see which sample (or subsample) any subsample was generated from. So in my example I chose to generate further subsamples from lineage “K” for space reasons, but you can see when we get to the 3rd level of subsampling (sequencing) I can still tell that these sequences were generated from RNA extraction #1, PCR#2.
  5. It would be nice if we could give subsamples nicknames… this would allow us to generate descriptive names in association with the machine generated sample name (thinking for sequences or PCR samples we could give nicknames that provide more experimental based info).
rintoult commented 2 years ago

We have spent the last two weeks, in my mind, but the rest of you before that discussing incrementation of material sample versions, that is only 1/2 of the problem. Has there already been discussion of incrementation of the specimen identifiers themselves?

ie: I receive a deposit of 20 new accessions, I need to assign them identifiers, I want them to be unique and sequential after the last identifier i used. I don't see how this could be a separate action under the current naming? But i would love to be shown!

rintoult commented 2 years ago

And Satpal - I added Emily - could you add her to all these issues going forward?

Thanks, Tara