dynverse / dyngen

Simulating single-cell data using gene regulatory networks 📠
https://dyngen.dynverse.org
Other
73 stars 6 forks source link

Simulate differing kinetics #15

Closed LouiseDck closed 5 years ago

LouiseDck commented 5 years ago

Adds random noise to kinetics for each simulation.

rcannood commented 5 years ago

Hey Louise!

Thanks for making this PR! Wouter or I will look at your contribution ASAP.

For now, I can see travis is complaining:

* checking for code/documentation mismatches ... WARNING
Codoc mismatches from documentation object 'generate_cells':
simulation_default
  Code: function(burn_time = 2, total_time = 10, ssa_algorithm =
                 ssa_etl(tau = 0.001), census_interval = 0.01,
                 experiment_params =
                 bind_rows(simulation_type_wild_type(num_simulations =
                 32), simulation_type_knockdown(num_simulations = 0)),
                 store_grn = TRUE, store_reaction_firings = FALSE,
                 store_reaction_propensities = FALSE,
                 kinetic_noise_strength = 0.1)

  Docs: function(burn_time = 2, total_time = 10, ssa_algorithm =
                 ssa_etl(tau = 0.001), census_interval = 0.01,
                 experiment_params =
                 bind_rows(simulation_type_wild_type(num_simulations =
                 32), simulation_type_knockdown(num_simulations = 0)),
                 store_grn = TRUE, store_reaction_firings = FALSE,
                 store_reaction_propensities = FALSE)

It tells you that the documentation does not match the actual code. Probably, you need to rerun roxygen, which generates documentation in man/*.Rd coming from the roxygen comments that are in the R/*.R files. Could you run roxygen (Ctrl+Shift+D, I think it is) and push the changes?

Robrecht

codecov[bot] commented 5 years ago

Codecov Report

Merging #15 into devel will decrease coverage by 0.01%. The diff coverage is 5.08%.

Impacted file tree graph

@@           Coverage Diff            @@
##           devel     #15      +/-   ##
========================================
- Coverage   7.37%   7.36%   -0.02%     
========================================
  Files         20      20              
  Lines       1980    2010      +30     
========================================
+ Hits         146     148       +2     
- Misses      1834    1862      +28
Impacted Files Coverage Δ
R/5_kinetics.R 5.71% <0%> (-0.54%) :arrow_down:
R/7_simulation.R 7.35% <8.57%> (+0.41%) :arrow_up:

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 780a195...b2f4e65. Read the comment docs.

rcannood commented 5 years ago

Thanks @LouiseDck ! I made a few small changes; most notably the parameter was changed from the sd to a function, which allows changing the noise to anything you want. I also added noise not just to strength, but also basal, wpr, wsr, xdr, ypr, ydr, and cooperativity. Afterwards, the k and max_protein values are recalculated.