search

Benchmarking-Initiative / Benchmark-Models-PEtab

A collection of mathematical models with experimental data in the PEtab format as benchmark problems in order to evaluate new and existing methodologies for data-based modelling
BSD 3-Clause "New" or "Revised" License
26 stars 14 forks source link

Add Lang_PLOSComputBiol2024 #210

Closed m-philipps closed 3 months ago

m-philipps commented 3 months ago

Add the model from Reusable rule-based cell cycle model explains compartment-resolved dynamics of 16 observables in RPE-1 cells (link to publication) in PR #202 .

paulflang commented 3 months ago

Hi Maren, thank you very much for helping with this PR. Here are the answers to your points. Please let me know where further clarification is needed.

A brief model description (one or two sentences)

The model describes the mammalian cell cycle, including the restriction point, G1/S, G2/M and metaphase/anaphase transitions.

A brief data description (one or two sentences)

Stallaert et al. (Cell Systems, 2021) obtained snapshot measurements of (phospho)protein abundance in a proliferating RPE1 cell population through iterative indirect immunofluorescence imaging. Using reCAT (Liu et al., Nature Communications, 2017), we ranked these snapshot measurements based on molecular time, from which we calculated wall-time via ERA (Kafti et al., Nature, 2013).

Differences between the implementation and the original publication are described

There are no semantic differences compared to the original publication. However, the PEtab files were reformatted to adhere to the requirements of this repository.

Experience of fitting / uncertainty analysis (e.g. optimizer used, hyperparameters, reproducibility of best fit)

We used saCeSS+parPE as hybrid global/local optimizer with self-adaptive hyperparameters and DHC as local solver option. Several modifications to the optimization problem (e.g. adding an offset parameter, adjusting the bounds) were needed to obtain the final optimization problem. Due to known parameter unidentifiability and high compute costs, we neither performed uncertainty analysis nor tried to reproduce the optimization results.

Source of nominal parameters (e.g.: taken from the original publication, or from your own fitting)

The nominalValues are taken from our own fitting as published in the original publication.

dilpath commented 2 months ago

Some small changes to the parameter table were made in #228, but nothing that should change the parameter estimation problem up to numerical noise.