Here is the original:

pubs.aip.org/aip/jcp/arti...

One of my favorites by Voltaire in describing Canada
“quelques arpents de neige, habités par des barbares, des ours et des castors” 🙃

Thanks for the link to your paper!

6/6
If you plate, you need REPOP.

Software -- github.com/PessoaP/REPOP
Preprint -- elifesciences.org/reviewed-pre...

Special thanks to the Lab Members @pedropessoaphd.bsky.social, Carol Lu and Stanimir Tashev

As well as Rory Kruithoff and @dpshepherd.bsky.social
#Biophysics #QuantitativeBiology

5/6
This is why we built REPOP, an #opensource tool to REconstruct POpulations from Plates.

Straightforward to use and with tutorials available on #GitHub

github.com/PessoaP/REPOP

With all the #Bayesian rigor and #PyTorch speed

4/6
As we show in the paper, this
- Overestimatese variability
- Can miss real structure in your population: Subpopulations and/or multimodality as biological differences across samples,

3/6
This assumes:
– No randomness in how many bacteria end up on the plate
– No randomness in the original swab

In reality, every step is noisy.

2/6
Plate counting is a simple:

You dilute a sample, plate a small volume, and count colonies.

Say you dilute by 200×, and count 50 colonies.
Easy just multiply 50 × 200 = 10k bacteria, right?

NOT QUITE...

3/3
In it, we simulate some general physical systems that violate the HMM's assumptions and demonstrate contradictory results that can arise. Surprisingly, the problems with HMM analysis only grow with better data acquisition (higher data acquisition rate and/or reduced noise).

2/3
HMM are classic in time series analysis, but they can yield confusing, seemingly contradictory results. In particular, when applying HMMs to physical systems where two key HMM assumptions, that state spaces are discrete, and that transitions are instantaneous, don't apply.