Barrick Lab :: ProtocolsGrowthRates

<noautolink>
<div style="text-align:justify">
---+ Growth Rates using "R studio"  <img src="%ATTACHURLPATH%/RStudio-Ball.png" alt="RStudio-Ball.png" width='50' height='50' /> - Overview                            

Determining growth rates for many cultures or from plate reader data can be a daunting task.  To automate this process, we've found that the Grofit package for R can be very useful and reliable.

---++ Download and install "R" and "R studio"

You'll need to download and install R on your computer.  Once R is installed, you can then download and install "R studio" which is easier to work with. 

You can download "R" here:  https://cran.r-project.org/mirrors.html<br/>
You can download "R studio" here: https://www.rstudio.com/

---++ Growing cells in plate reader and measuring OD600

All strains tested should at least have 3 replicates.  Due to small variations in temperature throughout a 96-well plate, best results will be obtained if replicates are randomly located at different wells.
Set up 200µL cultures with 2µL from a pre-conditioned (same media as actual test) overnight culture.  These 100-fold dilutions should be good enough for OD600 measurements that will give out reliable information for lag phase, exponential phase and stationary phase. To allow cultures to aerate properly, set the plate reader to shake (1mm radious) at least every 5 minutes; normally this would be before and after taking reads. 

A good 96-well plates to are clear Costar brand.  Make sure you remove the lid before inserting in the plate reader, we've had no contamination (as seen on wells with LB blanks) or problems with this.  

---++ Converting your file with OD data to the proper format
What is most critical is the format and file type of the input data file containing the OD600 values.  Most plate readers produce an Excel worksheet with the data results.  With such a file, the first thing that must be done is to convert it to a comma delimited file.  This is simply done by using "save as" in Excel and selecting *.csv as the file format.
Your file should look similar to the one provided as example here: 
   * [[%ATTACHURL%/yourdatafile.csv][yourdatafile.csv]]: Example spreadsheet of your OD600 data file

---++ Create a "times only" csv file
Grofit needs an additional file that will only have the times (e.g., 10, 20, 30 mins ...) for every time point measurement. 
An example of this file is given here:
   * [[%ATTACHURL%/timesonlyworksheet.csv][timesonlyworksheet.csv]]: Times only worksheet for Grofit
It is critical that the number of time points in this file, match the number of OD readings on your data file.

---++ Calculating Growth Rates using Grofit R package 
The following is a short script (by Julie Perreau) on "R studio" that performs the downloading, installing, running and printing out results. You execute each line in the script by pressing Ctrl+Enter.  For each one, you must check no errors are given. You can use the example timesonlyworksheet.csv and yourdatafile.csv files provided above as a test. 

setwd("C:/.../R_GrowthData")          *#sets the working directory to the folder (here named R_GrowthData) you'll be working on your computer*   <br/>
install.packages(c("grofit", "tidyr", "reshape2"))<br/>
library(grofit)<br/>
library(tidyr)<br/>
library(reshape2)<br/>
<br/>
growthdata <- read.csv("yourdatafile.csv",sep=",", header=TRUE, check.names = FALSE)  <br/>             
timedata <- read.csv("timesonlyworksheet.csv", sep=",", header=TRUE, check.names = FALSE) <br/>
gro <- grofit(timedata, growthdata)                    *#runs grofit  <br/>*
<br/>
summary_table<-summary.gcFit(gro$gcFit)       *#makes a summary table*   <br/>
write.csv(summary_table,"FinalTable.csv")         *#makes a csv file of the summary table*   <br/>

---+++ Accepting/declining model fits:
Once Grofit is run, you will be prompted to answer yes (y) or no (n) to accept or decline the model fits generated for each individual well.  Most likely, all you need to do is say yes to *all* of them, unless there is some obvious model fit mistake which would make some sample unreliable.  It is normal for Grofit to be unable to fit the curves to one of the 4 models tried, thus the ERROR sometimes given is not going to affect the final outcome, it just means it couldn't use one of the models to determine the best fit.
<br/>
     <img src="%ATTACHURLPATH%/Satisfied.JPG" alt="Satisfied.JPG" width='880' height='402' />

---+++ Reading summary table results:

The table summary file produced by Grofit will give you all the parameter values (mu=GrowthRates, lambda=LagPhaseTime and A=MaxAbsorbance) generated from the best model fits.  Grofit uses 4 possible models; see documentation for more information. 


---++ References

   1 https://www.jstatsoft.org/article/view/v033i07
   1 https://cran.r-project.org/web/packages/grofit/index.html

---++ *Contributors*

   * Gabriel Suárez
   * Julie Perreau

-- Main.GabrielSuarez - 15 Dec 2016

Attachments

Topic attachments
I	Attachment	History	Action	Size	Date	Who	Comment
png	RStudio-Ball.png	r1	manage	149.1 K	2016-12-15 - 23:38	GabrielSuarez
jpg	Satisfied.JPG	r1	manage	110.9 K	2016-12-15 - 23:16	GabrielSuarez	Accepting or declining model fit for each sample curve
csv	timesonlyworksheet.csv	r1	manage	58.5 K	2016-12-15 - 22:57	GabrielSuarez	Times only worksheet for Grofit
csv	yourdatafile.csv	r1	manage	159.8 K	2016-12-15 - 23:01	GabrielSuarez	Example spreadsheet of your OD600 data file

This topic: Lab > WebHome > ProtocolList > ProtocolsGrowthRates
Topic revision: r3 - 2016-12-15 - GabrielSuarez