Compare Parameters

Table of Contents


Basic usage

Users can directly compare growth parameters for two samples or two conditions using compare.py. This assumes that you have already analyzed your samples by pooling with AMiGA. See How to infer summary statistics for pooled replicates?. For example, let’s say you ran the following command:

python $amiga/fit.py -i /Users/firasmidani/experiment/ -o CD2015 --pool-by "Substrate,Isolate" --sample-posterior

The --sample-posterior argument asks AMiGA to compute summary statistics for growth parameters (i.e., mean and standard deviations). Next, the following command will compare the growth of the CD2015 isolate in PM 1 on fructose and trehalose. It will generate the following table.

python $amiga/compare.py -i /Users/firasmidani/experiment/summary/CD2015_summary.txt -o CD2015_Fructose_vs_Trehalose -s "Substrate:D-Fructose,D-Trehalose;Isolate:CD2015;PM:1" --confidence 95
  • -i must point to the summary file genearted by AMiGA.
  • -o assigns a filename for the results, otherwise, the filename will be a unique time stamp.
  • -s susbetting arguments must reduce analysis to only two conditions, otherwise, this command will fail and result in an error.
  • --confidence allows you to change the magnitude of the confidence interval. The default is 95.


Typical output

This will generate the below which will be saved in the same directory as the input file. Keep in mind that the below example is based only on two technical replicates for each condition, so the statistical power is pretty low, but the differences in growth dynamics are pretty clear. See Figure 2 of AMiGA manuscript for growth curves.


Substrate D-Trehalose D-Fructose D-Trehalose D-Fructose  
PM 1 1 1 1  
Isolate CD2015 CD2015 CD2015 CD2015  
Parameter Mean Mean 95.0% CI 95.0% CI Sig. Diff.
AUC (log) 36.435 36.218 [35.933,36.938] [35.888,36.547] FALSE
Death (log) 0.095 1.427 [-0.037,0.226] [1.335,1.519] TRUE
Diauxie 0 0 NA NA FALSE
Death Rate -0.041 -0.307 [-0.106,0.023] [-0.344,-0.270] TRUE
Growth Rate 0.403 0.604 [0.331,0.476] [0.571,0.637] TRUE
Carrying Capacity (log) 2.125 2.259 [2.076,2.173] [2.219,2.298] TRUE
Lag Time 0.474 1.037 [0.187,0.761] [0.929,1.145] TRUE
Adaptation Time 0.167 0.475 [0.167,0.167] [0.195,0.755] TRUE
Time at Max. Death Rate 20.817 19.723 [18.147,23.486] [18.831,20.615] FALSE
Time at Max. Growth Rate 0.187 2.62 [0.005,0.368] [2.417,2.823] TRUE
Time at Carrying Capacity 17.492 9.967 [13.882,21.102] [9.408,10.526] TRUE
Doubling Time 1.732 1.148 [1.419,2.046] [1.086,1.211] TRUE


Comparing growth curves analyzed in different summary files

Let’s say you want to compare two samples but their growth summary are in different files. You can manually create a new summary file with only the two samples (rows) that you are interested in anlayzing, then passing this file to the -i argument. But you can also pass multiple -i arguments to AMiGA. The below will find the growth summary for CD2015 and CD1007 on fructose the compare them against each other.

python $amiga/compare.py -i /Users/firasmidani/experiment/summary/CD2015_summary.txt -i /Users/firasmidani/experiment/summary/CD1007_summary.txt -o CD2015_vs_CD1007_on_Fructose -s 'Substrate:D-Fructose;Isolate:CD2015,CD1007' --confidence 95


Command-line arguments

To see the full list of arguments that amiga compare will accept, run

python amiga.py compare --help

which will return the following message

usage: amiga.py [-h] -i INPUT -o OUTPUT -s SUBSET [--confidence CONFIDENCE]
                [--verbose]

Compare two growth curves

optional arguments:
  -h, --help            show this help message and exit
  -i INPUT, --input INPUT
  -o OUTPUT, --output OUTPUT
  -s SUBSET, --subset SUBSET
  --confidence CONFIDENCE
                        Must be between 80 and 100. Default is 95.
  --verbose


See more details for these arguments in Command Line Interface