omniplate.Omniplate.platereader#

class platereader(dnames=False, anames=False, wdir='.', platereadertype='Tecan', dsheetnumbers=False, asheetnumbers=False, ODfname=None, info=True, ls=True)[source]#

Bases: object

for analyzing plate-reader data, correcting for autofluorescence, and determining growth rates.

All data is stored used Panda’s dataframes and plotted using Seaborn.

Three dataframes are created. If p is an instance of the platereader class, then p.r contains the raw data for each well in the plate; p.s contains the processed time-series using the data from all relevant wells; and p.sc constains any summary statistics, such as ‘max gr’.

For time series sampled from a Gaussian process, the mean is used as the statistics and errors are estimated by the standard deviation. For statistics calculated from time series, the median is used and errors are estimated by half the interquartile range, with the distribution of the statistic calculated by sampling time series.

A typical work flow is:

>>> import omniplate as om

then either

>>> p= om.platereader('GALdata.xlsx', 'GALcontents.xlsx',
...                    wdir= 'data/')

or

>>> p= om.platereader()
>>> p.load('GALdata.xls', 'GALcontents.xlsx')

and to analyse OD data

>>> p.plot('OD', plate= True)
>>> p.correctOD()
>>> p.correctmedia()
>>> p.plot(y= 'OD')
>>> p.plot(y= 'OD', hue= 'strain',
...        conditionincludes= ['Gal', 'Glu'],
...        strainexcludes= 'HXT7')
>>> p.getstats('OD')

and for fluorescence data

>>> p.correctauto(['GFP', 'AutoFL'])
>>> p.plot(y= 'c-GFPperOD', hue= 'condition')

and to save the results

>>> p.savefigs()
>>> p.exportdf()

General properties of the data and of previous processing are shown with:

>>> p.info
>>> p.attributes()
>>> p.corrections()
>>> p.log

See also

for a tutorial, which can be opened directly using

>>> p.webhelp()
Attributes
attributes

Displays the names of the attributes of the current instance of platereader and acts as a check to see what variables have been calculated or determined.

close

Close all figures.

info

Displays conditions, strains, and datatypes.

log

Prints a log of all methods called and their arguments.

ls

List all files in the working directory.

Methods

add_to_sc([newcolumn, s_column, func, ...])

Applies func to a column in the s dataframe and stores the results in the sc dataframe.

addcolumn(newcolumnname, oldcolumn, ...)

Adds a new column to all dataframes by parsing an existing column.

addcommonvar([var, dvar, varmin, varmax, ...])

Adds to time-dependent dataframes a common variable whose values only come from a fixed array so that they are from the same array for all experiments.

addnumericcolumn(newcolumnname, oldcolumn[, ...])

Adds a new numeric column by parsing the numbers from the entries of an existing column.

averageoverexpts(condition, strain[, tvr, ...])

Uses a Matern Gaussian process to average a time-dependent variable over all experiments.

changewdir(wdir)

Change working directory.

contentsofwells(wlist)

Displays contents of wells

correctOD([figs, odmatch, experiments, ...])

Corrects OD data for the non-linear relationship between OD and cell number.

correctauto([f, refstrain, figs, ...])

Corrects fluorescence data for autofluorescence by comparing with the fluorescence of an untagged reference strain.

corrections([experiments, conditions, ...])

Displays the status of corrections made for the specified strains, conditions, and experiments.

correctmedia([datatypes, commonmedia, ...])

Corrects OD or fluorescence for the OD or fluorescence of the media using data from wells marked Null.

exportdf([commonname, type])

Exports the dataframes as either tab-delimited or csv or json files.

getdataframe([dfname, experiments, ...])

Obtain a subset of the data in a dataframe, which can be used plotting directly.

getstats([dtype, bd, cvfn, ...])

Calls fitderiv.py to estimate the first and second time-derivatives of, typically, OD using a Gaussian process (Swain et al., 2016) and find corresponding summary statistics.

ignorewells([exclude, experiments, ...])

Allows wells to be ignored in any future processing.

importdf(commonnames[, info, sep])

Import dataframes saved as either json or csv or tsv files.

load(dnames[, anames, platereadertype, ...])

Loads raw data files generated by the plate reader and the corresponding annotation files.

plot([x, y, hue, style, size, kind, col, ...])

Plots from the underlying dataframes (chosen automatically) using Seaborn's relplot, which is described at https://seaborn.pydata.org/generated/seaborn.relplot.html

rename(translatedict)

Uses a dictionary to replace all occurrences of a strain or a condition with an alternative.

restricttime([tmin, tmax])

Restrict the processed data to a range of time, ignoring points outside this time range.

savefigs([fname, onefile])

Saves all current figures to PDF, either to one file or each to a separate file.

savelog([fname])

Save log to file.

showwells([concise, sortby, experiments, ...])

Displays wells for specified experiments, conditions, and strains.

webhelp([browser])

Opens detailed examples of how to use in omniplate in a web browser.

Initiate and potentially immediately load data for processing.

Parameters
dnames: string or list of strings, optional

The name of the file containing the data from the plate reader or a list of file names.

anames: string or list of strings, optional

The name of file containing the corresponding annotation or a list of file names.

wdir: string, optional

The working directory where the data files are stored and where output will be saved.

platereadertype: string

The type of plate reader, currently either ‘Tecan’ or ‘Sunrise’ or ‘old Tecan’.

dsheetnumbers: integer or list of integers, optional

The relevant sheets of the Excel files storing the data.

asheetnumbers: integer or list of integers, optional

The relevant sheets of the corresponding Excel files for the annotation.

ODfname: string, optional

The name of the file with the dilution data used to correct OD for its non-linear dependence on numbers of cells. If unspecified, data for haploid budding yeast growing in glucose is used.

info: boolean

If True (default), display summary information on the data once loaded.

ls: boolean

If True (default), display contents of working directory.

Attributes
attributes

Displays the names of the attributes of the current instance of platereader and acts as a check to see what variables have been calculated or determined.

close

Close all figures.

info

Displays conditions, strains, and datatypes.

log

Prints a log of all methods called and their arguments.

ls

List all files in the working directory.

Methods

add_to_sc([newcolumn, s_column, func, ...])

Applies func to a column in the s dataframe and stores the results in the sc dataframe.

addcolumn(newcolumnname, oldcolumn, ...)

Adds a new column to all dataframes by parsing an existing column.

addcommonvar([var, dvar, varmin, varmax, ...])

Adds to time-dependent dataframes a common variable whose values only come from a fixed array so that they are from the same array for all experiments.

addnumericcolumn(newcolumnname, oldcolumn[, ...])

Adds a new numeric column by parsing the numbers from the entries of an existing column.

averageoverexpts(condition, strain[, tvr, ...])

Uses a Matern Gaussian process to average a time-dependent variable over all experiments.

changewdir(wdir)

Change working directory.

contentsofwells(wlist)

Displays contents of wells

correctOD([figs, odmatch, experiments, ...])

Corrects OD data for the non-linear relationship between OD and cell number.

correctauto([f, refstrain, figs, ...])

Corrects fluorescence data for autofluorescence by comparing with the fluorescence of an untagged reference strain.

corrections([experiments, conditions, ...])

Displays the status of corrections made for the specified strains, conditions, and experiments.

correctmedia([datatypes, commonmedia, ...])

Corrects OD or fluorescence for the OD or fluorescence of the media using data from wells marked Null.

exportdf([commonname, type])

Exports the dataframes as either tab-delimited or csv or json files.

getdataframe([dfname, experiments, ...])

Obtain a subset of the data in a dataframe, which can be used plotting directly.

getstats([dtype, bd, cvfn, ...])

Calls fitderiv.py to estimate the first and second time-derivatives of, typically, OD using a Gaussian process (Swain et al., 2016) and find corresponding summary statistics.

ignorewells([exclude, experiments, ...])

Allows wells to be ignored in any future processing.

importdf(commonnames[, info, sep])

Import dataframes saved as either json or csv or tsv files.

load(dnames[, anames, platereadertype, ...])

Loads raw data files generated by the plate reader and the corresponding annotation files.

plot([x, y, hue, style, size, kind, col, ...])

Plots from the underlying dataframes (chosen automatically) using Seaborn's relplot, which is described at https://seaborn.pydata.org/generated/seaborn.relplot.html

rename(translatedict)

Uses a dictionary to replace all occurrences of a strain or a condition with an alternative.

restricttime([tmin, tmax])

Restrict the processed data to a range of time, ignoring points outside this time range.

savefigs([fname, onefile])

Saves all current figures to PDF, either to one file or each to a separate file.

savelog([fname])

Save log to file.

showwells([concise, sortby, experiments, ...])

Displays wells for specified experiments, conditions, and strains.

webhelp([browser])

Opens detailed examples of how to use in omniplate in a web browser.
__init__(dnames=False, anames=False, wdir='.', platereadertype='Tecan', dsheetnumbers=False, asheetnumbers=False, ODfname=None, info=True, ls=True)[source]#

Initiate and potentially immediately load data for processing.

Parameters
dnames: string or list of strings, optional

The name of the file containing the data from the plate reader or a list of file names.

anames: string or list of strings, optional

The name of file containing the corresponding annotation or a list of file names.

wdir: string, optional

The working directory where the data files are stored and where output will be saved.

platereadertype: string

The type of plate reader, currently either ‘Tecan’ or ‘Sunrise’ or ‘old Tecan’.

dsheetnumbers: integer or list of integers, optional

The relevant sheets of the Excel files storing the data.

asheetnumbers: integer or list of integers, optional

The relevant sheets of the corresponding Excel files for the annotation.

ODfname: string, optional

The name of the file with the dilution data used to correct OD for its non-linear dependence on numbers of cells. If unspecified, data for haploid budding yeast growing in glucose is used.

info: boolean

If True (default), display summary information on the data once loaded.

ls: boolean

If True (default), display contents of working directory.

Methods

__init__([dnames, anames, wdir, ...])

Initiate and potentially immediately load data for processing.

add_to_sc([newcolumn, s_column, func, ...])

Applies func to a column in the s dataframe and stores the results in the sc dataframe.

addcolumn(newcolumnname, oldcolumn, ...)

Adds a new column to all dataframes by parsing an existing column.

addcommonvar([var, dvar, varmin, varmax, ...])

Adds to time-dependent dataframes a common variable whose values only come from a fixed array so that they are from the same array for all experiments.

addnumericcolumn(newcolumnname, oldcolumn[, ...])

Adds a new numeric column by parsing the numbers from the entries of an existing column.

averageoverexpts(condition, strain[, tvr, ...])

Uses a Matern Gaussian process to average a time-dependent variable over all experiments.

changewdir(wdir)

Change working directory.

contentsofwells(wlist)

Displays contents of wells

correctOD([figs, odmatch, experiments, ...])

Corrects OD data for the non-linear relationship between OD and cell number.

correctauto([f, refstrain, figs, ...])

Corrects fluorescence data for autofluorescence by comparing with the fluorescence of an untagged reference strain.

corrections([experiments, conditions, ...])

Displays the status of corrections made for the specified strains, conditions, and experiments.

correctmedia([datatypes, commonmedia, ...])

Corrects OD or fluorescence for the OD or fluorescence of the media using data from wells marked Null.

exportdf([commonname, type])

Exports the dataframes as either tab-delimited or csv or json files.

getdataframe([dfname, experiments, ...])

Obtain a subset of the data in a dataframe, which can be used plotting directly.

getstats([dtype, bd, cvfn, ...])

Calls fitderiv.py to estimate the first and second time-derivatives of, typically, OD using a Gaussian process (Swain et al., 2016) and find corresponding summary statistics.

ignorewells([exclude, experiments, ...])

Allows wells to be ignored in any future processing.

importdf(commonnames[, info, sep])

Import dataframes saved as either json or csv or tsv files.

load(dnames[, anames, platereadertype, ...])

Loads raw data files generated by the plate reader and the corresponding annotation files.

plot([x, y, hue, style, size, kind, col, ...])

Plots from the underlying dataframes (chosen automatically) using Seaborn's relplot, which is described at https://seaborn.pydata.org/generated/seaborn.relplot.html

rename(translatedict)

Uses a dictionary to replace all occurrences of a strain or a condition with an alternative.

restricttime([tmin, tmax])

Restrict the processed data to a range of time, ignoring points outside this time range.

savefigs([fname, onefile])

Saves all current figures to PDF, either to one file or each to a separate file.

savelog([fname])

Save log to file.

showwells([concise, sortby, experiments, ...])

Displays wells for specified experiments, conditions, and strains.

webhelp([browser])

Opens detailed examples of how to use in omniplate in a web browser.

Attributes

attributes

Displays the names of the attributes of the current instance of platereader and acts as a check to see what variables have been calculated or determined.

close

Close all figures.

info

Displays conditions, strains, and datatypes.

log

Prints a log of all methods called and their arguments.

ls

List all files in the working directory.
add_to_sc(newcolumn=None, s_column=None, func=None, experiments='all', experimentincludes=False, experimentexcludes=False, conditions='all', conditionincludes=False, conditionexcludes=False, strains='all', strainincludes=False, strainexcludes=False)[source]#

Applies func to a column in the s dataframe and stores the results in the sc dataframe.

Parameters
newcolumn: string

The name of the new column in the sc dataframe

s_column: string

The name of the column in s dataframe from which the data is to be processed

func: function

The function to be applied to the data in the s dataframe.

Examples

>>> p.add_to_sc(newcolumn= "max GFP", s_column= "GFP mean",
...             func= np.nanmax)
>>> p.add_to_sc(newcolumn= "GFP lower quartile", s_column= "GFP mean",
...             func= lambda x: np.nanquantile(x, 0.25))
addcolumn(newcolumnname, oldcolumn, newcolumnvalues)[source]#

Adds a new column to all dataframes by parsing an existing column. All possible entries for the new column are specified as strings and the entry in the new column will be whichever of these strings is present in the entry of the existing column.

Parameters
newcolumnname: string

The name of the new column.

oldcolumn: string

The name of the column to be parsed to create the new column.

newcolumnvalues: list of strings

All of the possible values for the entries in the new column.

addcommonvar(var='time', dvar=None, varmin=None, varmax=None, figs=True, experiments='all', experimentincludes=False, experimentexcludes=False, conditions='all', conditionincludes=False, conditionexcludes=False, strains='all', strainincludes=False, strainexcludes=False)[source]#

Adds to time-dependent dataframes a common variable whose values only come from a fixed array so that they are from the same array for all experiments. This common variable allows averaging across experiments and typically is time.

For example, the plate reader often does not perfectly increment time between measurements and different experients can have slightly different time points despite the plate reader having the same settings. These unique times prevent seaborn from taking averages.

If experiments have measurements that start at the same time point and have the same interval between measurements, then setting a commontime for all experiments will allow seaborn to perform averaging.

The array of the common variable runs from varmin to varmax with an interval dvar. These parameters are automatically calculated, but may be specified.

Each instance of var is assigned a common value - the closest instance of the common variable to the instance of var. Measurements are assumed to the same for the true instance of var and for the assigned common value, which may generate errors if these two are sufficiently distinct.

An alternative method is averageoverexpts.

Parameters
var: string

The variable from which the common variable is generated, typically ‘time’.

dvar: float, optional

The interval between the values comprising the common array.

varmin: float, optional

The minimum of the common variable.

varmax: float, optional

The maximum of the common variable.

figs: boolean

If True, generate plot to check if the variable and the common variable generated from it are sufficiently close in value.

experiments: string or list of strings

The experiments to include.

conditions: string or list of strings

The conditions to include.

strains: string or list of strings

The strains to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

conditionincludes: string, optional

Selects only conditions that include the specified string in their name.

conditionexcludes: string, optional

Ignores conditions that include the specified string in their name.

strainincludes: string, optional

Selects only strains that include the specified string in their name.

strainexcludes: string, optional

Ignores strains that include the specified string in their name.

addnumericcolumn(newcolumnname, oldcolumn, picknumber=0, leftsplitstr=None, rightsplitstr=None, asstr=False)[source]#

Adds a new numeric column by parsing the numbers from the entries of an existing column. It is best to run this command only after the basic analyses - ignorewells, correctOD, and correctmedia - have been performed because it changes the structure of the dataframes and may cause errors.

Parameters
newcolumnname: string

The name of the new column.

oldcolumn: string

The name of column to be parsed.

picknumber: integer

The number to pick from the list of numbers extracted from the existing column’s entry.

leftsplitstr: string, optional

Split the entry of the column using whitespace and parse numbers from the substring to the immediate left of leftsplitstr rather than the whole entry.

rightsplitstr: string, optional

Split the entry of the column using whitespace and parse numbers from the substring to the immediate right of rightsplitstr rather than the whole entry.

asstr: boolean

If True, convert the numeric value to a string to improve plots with seaborn.

Examples

To extract concentrations from conditions use

>>> p.addnumericcolumn('concentration', 'condition')

For a condition like ‘0.5% Raf 0.05ug/mL Cycloheximide’, use

>>> p.addnumericcolumn('raffinose', 'condition',
...                     picknumber= 0)
>>> p.addnumericcolumn('cycloheximide', 'condition',
...                     picknumber= 1)
property attributes#

Displays the names of the attributes of the current instance of platereader and acts as a check to see what variables have been calculated or determined.

averageoverexpts(condition, strain, tvr='OD mean', bd=False, addnoise=True, plot=False)[source]#

Uses a Matern Gaussian process to average a time-dependent variable over all experiments.

An alternative and best first choice is to use addcommonvar.

Parameters
condition: string

The condition of interest.

strain: string

The strain of interest.

tvr: float

The time-dependent variable to be averaged. For example, ‘c-GFPperOD’ or ‘OD mean’.

bd: dictionary, optional

The limits on the hyperparameters for the Matern Gaussian process. For example, {0: (-5,5), 1: (-4,4), 2: (-5,2)} where the first element controls amplitude, setting the bounds to 1e-5 and 1e5, the second controls flexibility, and the third determines the magnitude of the measurement error.

addnoise: boolean

If True, add the fitted magnitude of the measurement noise to the predicted standard deviation for better comparison with the spread of the data.

Returns
res: dictionary

{‘t’ : time, tvr : time-dependent data, ‘mn’ : mean, ‘sd’ : standard deviation} where ‘mn’ is the average found and ‘sd’ is its standard deviation. ‘tvr’ is the data used to find the average.

Examples

>>> p.averageoverexpts('1% Gal', 'GAL2', bd= {1: [-1,-1])})
changewdir(wdir)[source]#

Change working directory.

Parameters
wdir: string

The new working directory specified from the current directory.

property close#

Close all figures.

contentsofwells(wlist)[source]#

Displays contents of wells

Parameters
wlist: string or list of string

Specifies the well or wells of interest.

Examples

>>> p.contentsofwells(['A1', 'E4'])
correctOD(figs=True, odmatch=0.3, experiments='all', experimentincludes=False, experimentexcludes=False, conditions='all', conditionincludes=False, conditionexcludes=False)[source]#

Corrects OD data for the non-linear relationship between OD and cell number. Requires a set of dilution data set, with the default being haploid yeast growing in glucose (collected by L Bandiera). An alternative can be loaded from a file - a txt file of two columns with OD specified in the first column and the dilution factor specified in descending order in the second.

figs: boolean, optional

If True, a plot of the fit to the dilution data is produced.

odmatch: float, optional

If non-zero, then the corrected OD is rescaled to equal the measured OD at this value. Only large ODs typically need to be corrected.

experiments: string or list of strings

The experiments to include.

conditions: string or list of strings

The conditions to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

conditionincludes: string, optional

Selects only conditions that include the specified string in their name.

conditionexcludes: string, optional

Ignores conditions that include the specified string in their name.

>>> p.correctOD()
>>> p.correctOD(figs= False)
correctauto(f=['GFP', 'AutoFL'], refstrain='WT', figs=True, experiments='all', experimentincludes=False, experimentexcludes=False, conditions='all', conditionincludes=False, conditionexcludes=False, strains='all', strainincludes=False, strainexcludes=False)[source]#

Corrects fluorescence data for autofluorescence by comparing with the fluorescence of an untagged reference strain.

The reference strain is used to estimate the autofluoresence via either the method of Licthen et al., 2014, where measurements of fluoescence at two wavelengths is required, or by using the fluorescence of the reference strain interpolated to the OD of the strain of interest (Berthoumieux et al., 2013).

Using two measurements of fluorescence is thought to be more accurate, particularly for low fluorescence measurements (Mihalcescu et al., 2015).

Arguments – f: string or list of strings

The fluorescence measurements, typically either [‘mCherry’] or [‘GFP’, ‘AutoFL’].

refstrain: string

The reference strain.

figs: boolean

If True, display plots showing the fits to the reference strain’s fluorescnce.

experiments: string or list of strings

The experiments to include.

conditions: string or list of strings

The conditions to include.

strains: string or list of strings

The strains to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

conditionincludes: string, optional

Selects only conditions that include the specified string in their name.

conditionexcludes: string, optional

Ignores conditions that include the specified string in their name.

strainincludes: string, optional

Selects only strains that include the specified string in their name.

strainexcludes: string, optional

Ignores strains that include the specified string in their name.

Notes

In principle

>>> p.correctmedia()

should be run before running correctauto when processing data with two fluorescence measurements.

It is unnecessary with only one fluorescence measurement because the normalisation is then done directly with the reference strain’s fluorescence and this fluorescence can include the fluorescence from the media.

In practice, running correctmedia may generate negative values of the fluorescence at some time points. These negative values will create NaNs in the corrected fluorescence, which are normally harmless.

With sufficiently many negative values of the fluorescence, however, correcting data with two fluorescence measurements can become corrupted.

If correctmedia generates negative fluorescence values, we therefore recommend comparing the corrected fluorescence between

>>> p.correctmedia()
>>> p.correctauto(['GFP', 'AutoFL')

and

>>> p.correctauto('GFP')

to determine if these negative values are deleterious.

References

S Berthoumieux, H De Jong, G Baptist, C Pinel, C Ranquet, D Ropers, J Geiselmann (2013). Shared control of gene expression in bacteria by transcription factors and global physiology of the cell. Mol Syst Biol, 9, 634.

CA Lichten, R White, IB Clark, PS Swain (2014). Unmixing of fluorescence spectra to resolve quantitative time-series measurements of gene expression in plate readers. BMC Biotech, 14, 1-11.

I Mihalcescu, MVM Gateau, B Chelli, C Pinel, JL Ravanat (2015). Green autofluorescence, a double edged monitoring tool for bacterial growth and activity in micro-plates. Phys Biol, 12, 066016.

Examples

To correct data with one type of fluorescence measurement, use:

>>> p.correctauto('GFP')
>>> p.correctauto('mCherry', refstrain= 'BY4741')

To correct data with two types of fluorescence measurement, use:

>>> p.correctauto(['GFP', 'AutoFL'])
>>> p.correctauto(['GFP', 'AutoFL'], refstrain= 'wild-type')
corrections(experiments='all', conditions='all', strains='all', experimentincludes=False, experimentexcludes=False, conditionincludes=False, conditionexcludes=False, strainincludes=False, strainexcludes=False)[source]#

Displays the status of corrections made for the specified strains, conditions, and experiments.

Parameters
experiments: string or list of strings

The experiments to include.

conditions: string or list of strings

The conditions to include.

strains: string or list of strings

The strains to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

conditionincludes: string, optional

Selects only conditions that include the specified string in their name.

conditionexcludes: string, optional

Ignores conditions that include the specified string in their name.

strainincludes: string, optional

Selects only strains that include the specified string in their name.

strainexcludes: string, optional

Ignores strains that include the specified string in their name.

Returns
df: dataframe

Contains the status of the corrections for the specified strains, conditions, and experiments.

Examples

>>> p.corrections()
>>> p.corrections(strainincludes= 'GAL')
correctmedia(datatypes='all', commonmedia=False, experiments='all', experimentincludes=False, experimentexcludes=False, conditions='all', conditionincludes=False, conditionexcludes=False, figs=False, log=True, frac=0.33)[source]#

Corrects OD or fluorescence for the OD or fluorescence of the media using data from wells marked Null. Uses lowess to smooth measurements of from all Null wells and subtracts this smoothed time series from the raw data.

Parameters
datatypes: string or list of strings

Data types to be corrected.

commonmedia: string

A condition containing Null wells that should be used to correct media for other conditions.

experiments: string or list of strings

The experiments to include.

conditions: string or list of strings

The conditions to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

conditionincludes: string, optional

Selects only conditions that include the specified string in their name.

conditionexcludes: string, optional

Ignores conditions that include the specified string in their name.

figs: boolean, optional

If True, display fits to data for the Null wells.

frac: float

The fraction of the data used for smoothing via lowess. https://www.statsmodels.org/devel/generated/statsmodels.nonparametric.smoothers_lowess.lowess.html

Examples

>>> p.correctmedia()
>>> p.correctmedia('OD')
>>> p.correctmedia(commonmedia= '1% Glu')
exportdf(commonname=False, type='tsv')[source]#

Exports the dataframes as either tab-delimited or csv or json files. Dataframes for the (processed) raw data, for summary data, and for summary statistics and corrections, as well as a log file, are exported.

Parameters
commonname: string, optional

The name used for the output files. If unspecified, the experiment or experiments is used.

type: string

The type of file for export, either ‘json’ or ‘csv’ or ‘tsv’.

Examples

>>> p.exportdf()
>>> p.exportdf('processed', type= 'json')
getdataframe(dfname='s', experiments='all', conditions='all', strains='all', experimentincludes=False, experimentexcludes=False, conditionincludes=False, conditionexcludes=False, strainincludes=False, strainexcludes=False, nonull=True)[source]#

Obtain a subset of the data in a dataframe, which can be used plotting directly.

dfname: string

The dataframe of interest either ‘r’ (raw data), ‘s’ (default; processed data), or ‘sc’ (summary statistics).

experiments: string or list of strings

The experiments to include.

conditions: string or list of strings

The conditions to include.

strains: string or list of strings

The strains to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

conditionincludes: string, optional

Selects only conditions that include the specified string in their name.

conditionexcludes: string, optional

Ignores conditions that include the specified string in their name.

strainincludes: string, optional

Selects only strains that include the specified string in their name.

strainexcludes: string, optional

Ignores strains that include the specified string in their name.

nonull: boolean, optional

If True, ignore ‘Null’ strains

Returns
ndf: dataframe
getstats(dtype='OD', bd=False, cvfn='matern', empirical_errors=False, noruns=10, exitearly=True, noinits=100, nosamples=100, logs=True, iskip=False, stats=True, figs=True, findareas=False, plotlocalmax=True, showpeakproperties=False, experiments='all', experimentincludes=False, experimentexcludes=False, conditions='all', conditionincludes=False, conditionexcludes=False, strains='all', strainincludes=False, strainexcludes=False, **kwargs)[source]#

Calls fitderiv.py to estimate the first and second time-derivatives of, typically, OD using a Gaussian process (Swain et al., 2016) and find corresponding summary statistics.

The derivatives are stored in the .s dataframe; summary statistics are stored in the .sc dataframe.

Parameters
dtype: string, optional

The type of data - ‘OD’, ‘GFP’, ‘c-GFPperOD’, or ‘c-GFP’ - for which the derivatives are to be found. The data must exist in the .r or .s dataframes.

bd: dictionary, optional

The bounds on the hyperparameters for the Gaussian process. For example, bd= {1: [-2,0])} fixes the bounds on the hyperparameter controlling flexibility to be 1e-2 and 1e0. The default for a Matern covariance function is {0: (-5,5), 1: (-4,4), 2: (-5,2)}, where the first element controls amplitude, the second controls flexibility, and the third determines the magnitude of the measurement error.

cvfn: string, optional

The covariance function used in the Gaussian process, either ‘matern’ or ‘sqexp’ or ‘nn’.

empirical_errors: boolean, optional

If True, measurement errors are empirically estimated from the variance across replicates at each time point and so vary with time. If False, the magnitude of the measurement error is fit from the data assuming that this magnitude is the same at all time points.

noruns: integer, optional

The number of attempts made for each fit. Each attempt is made with random initial estimates of the hyperparameters within their bounds.

exitearly: boolean, optional

If True, stop at the first successful fit. If False, use the best fit from all successful fits.

noinits: integer, optional

The number of random attempts to find a good initial condition before running the optimization.

nosamples: integer, optional

The number of samples used to calculate errors in statistics by bootstrapping.

logs: boolean, optional

If True, find the derivative of the log of the data and should be True to determine the specific growth rate when dtype= ‘OD’.

iskip: integer, optional

Use only every iskip’th data point to increase speed.

stats: boolean, optional

If False, do not calculate statistics.

figs: boolean, optional

If True, plot both the fits and inferred derivative.

findareas: boolean, optional

If True, find the area under the plot of gr vs OD and the area under the plot of OD vs time. Setting to True can make getstats slow.

plotlocalmax: boolean, optional

If True, mark the highest local maxima found, which is used to calculate statistics, on any plots.

showpeakproperties: boolean, optional

If True, show properties of any local peaks that have found by scipy’s find_peaks. Additional properties can be specified as kwargs and are passed to find_peaks.

experiments: string or list of strings

The experiments to include.

conditions: string or list of strings

The conditions to include.

strains: string or list of strings

The strains to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

conditionincludes: string, optional

Selects only conditions that include the specified string in their name.

conditionexcludes: string, optional

Ignores conditions that include the specified string in their name.

strainincludes: string, optional

Selects only strains that include the specified string in their name.

strainexcludes: string, optional

Ignores strains that include the specified string in their name.

kwargs: for scipy’s find_peaks

To set the minimum property of a peak. e.g. prominence= 0.1 and width= 15 (specified in numbers of x-points or y-points and not real units). https://docs.scipy.org/doc/scipy/reference/generated/scipy.signal.find_peaks.html

References

PS Swain, K Stevenson, A Leary, LF Montano-Gutierrez, IB Clark, J Vogel, T Pilizota. (2016). Inferring time derivatives including cell growth rates using Gaussian processes. Nat Commun, 7, 1-8.

Examples

>>> p.getstats()
>>> p.getstats(conditionincludes= 'Gal')
>>> p.getstats(noruns= 10, exitearly= False)

If the fits are poor, often changing the bounds on the hyperparameter for the measurement error helps:

>>> p.getstats(bd= {2: (-3,0)})
ignorewells(exclude=[], experiments='all', experimentincludes=False, experimentexcludes=False, clearall=False)[source]#

Allows wells to be ignored in any future processing. If called several times, the default behaviour is for any previously ignored wells not to be re-instated.

exclude: list of strings

List of labels of wells on the plate to be excluded.

experiments: string or list of strings

The experiments to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

clearall: boolean

If True, all previously ignored wells are re-instated.

importdf(commonnames, info=True, sep='\t')[source]#

Import dataframes saved as either json or csv or tsv files.

Parameters
commonnames: list of strings

A list of names for the files to be imported with one string for each experiment.

Examples

>>> p.importdf('Gal')
>>> p.importdf(['Gal', 'Glu', 'Raf'])
property info#

Displays conditions, strains, and datatypes.

load(dnames, anames=False, platereadertype='Tecan', dsheetnumbers=False, asheetnumbers=False, ODfname=None, info=True)[source]#

Loads raw data files generated by the plate reader and the corresponding annotation files.

Parameters
dnames: string or list of strings, optional

The name of the file containing the data from the plate reader or a list of file names.

anames: string or list of strings, optional

The name of file containing the corresponding annotation or a list of file names.

platereadertype: string

The type of plate reader, currently either ‘Tecan’ or ‘Sunrise’ or ‘old Tecan’.

dsheetnumbers: integer or list of integers, optional

The relevant sheets of the Excel files storing the data.

asheetnumbers: integer or list of integers, optional

The relevant sheets of the corresponding Excel files for the annotation.

ODfname: string, optional

The name of the file with the dilution data used to correct OD for its non-linear dependence on numbers of cells. If unspecified, data for haploid budding yeast growing in glucose is used.

info: boolean

If True (default), display summary information on the data once loaded.

Examples
——-
>>> p.load(‘Data.xlsx’, ‘DataContents.xlsx’)
>>> p.load(‘Data.xlsx’, ‘DataContents.xlsx’, info= False)
>>> p.load(‘Data.xlsx’, ‘DataContents.xlsx’,
… ODfname= ‘ODcorrection_Glucose_Diploid.txt’)
property log#

Prints a log of all methods called and their arguments.

property ls#

List all files in the working directory. A dictionary of available datasets is created as a shortcut.

plot(x='time', y='OD', hue='strain', style='condition', size=None, kind='line', col=None, row=None, height=5, aspect=1, ymin=None, figsize=False, returnfacetgrid=False, title=None, plate=False, wells=False, nonull=False, messages=False, sortby=False, experiments='all', conditions='all', strains='all', experimentincludes=False, experimentexcludes=False, conditionincludes=False, conditionexcludes=False, strainincludes=False, strainexcludes=False, **kwargs)[source]#

Plots from the underlying dataframes (chosen automatically) using Seaborn’s relplot, which is described at https://seaborn.pydata.org/generated/seaborn.relplot.html

Parameters
x: string

The variable - column of the dataframe - for the x-axis.

y: string

The variable - column of the dataframe - for y-axis.

hue: string

The variable whose variation will determine the colours of the lines plotted. From Seaborn.

style: string

The variable whose variation will determine the style of each line. From Seaborn.

size: string

The variable whose vairation will determine the size of each marker. From Seaborn.

kind: string

Either ‘line’ or ‘scatter’, which determines the type of plot. From Seaborn.

col: string, optional

The variable that varies over the columns in a multipanel plot. From Seaborn.

row: string, optional

The variable that varies over the rows in a multipanel plot. From Seaborn.

height: float, optional

The height of the individual panels in a multipanel plot. From Seaborn.

aspect: float, optional

The aspect ratio of the individual panels in a multipanel plot. From Seaborn.

ymin: float, optional

The minimum y-value

figsize: tuple, optional

A tuple of (width, height) for the size of figure. Ignored if wells= True or plate= True.

returnfacetgrid: boolean, optional

If True, return Seaborn’s facetgrid object created by relplot

title: float, optional

The title of the plot (overwrites any default titles).

plate: boolean, optional

If True, data for each well for a whole plate are plotted in one figure.

wells: boolean, optional

If True, data for the individual wells is shown.

nonull: boolean, optional

If True, ‘Null’ strains are not plotted.

sortby: list of strings, optional

A list of columns to sort the data in the dataframe and passed to pandas sort_values.

messsages: boolean, optional

If True, print warnings for any data requested but not found.

experiments: string or list of strings

The experiments to include.

conditions: string or list of strings

The conditions to include.

strains: string or list of strings

The strains to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

conditionincludes: string, optional

Selects only conditions that include the specified string in their name.

conditionexcludes: string, optional

Ignores conditions that include the specified string in their name.

strainincludes: string, optional

Selects only strains that include the specified string in their name.

strainexcludes: string, optional

Ignores strains that include the specified string in their name.

kwargs: for Seaborn’s relplot

https://seaborn.pydata.org/generated/seaborn.relplot.html

Returns
sfig: Seaborn’s facetgrid object generated by relplot if
returnfacetgrid= True

Examples

>>> p.plot(y= 'OD', plate= True)
>>> p.plot(y= 'OD', wells= True, strainincludes= 'Gal10:GFP')
>>> p.plot(y= 'OD')
>>> p.plot(x= 'OD', y= 'gr')
>>> p.plot(y= 'c-GFPperOD', nonull= True, ymin= 0)
>>> p.plot(y= 'c-GFPperOD', conditionincludes= '2% Mal',
...        hue= 'strain')
>>> p.plot(y= 'c-mCherryperOD', conditions= ['0.5% Mal',
...        '1% Mal'], hue= 'strain', style= 'condition',
...         nonull= True, strainincludes= 'mCherry')
>>> p.plot(y= 'c-GFPperOD', col= 'experiment')
>>> p.plot(y= 'max gr')
rename(translatedict)[source]#

Uses a dictionary to replace all occurrences of a strain or a condition with an alternative. Note that instances of self.progress will not be updated.

Parameters
translatedict: dictionary

A dictionary of old name - new name pairs

restricttime(tmin=None, tmax=None)[source]#

Restrict the processed data to a range of time, ignoring points outside this time range.

Note that data in the .s dataframe outside the time range is lost. Exporting the dataframes before running restricttime is recommended.

Parameters
tmin: float

The minimum value of time, with data kept only for t >= tmin.

tmax: float

The maximum value of time, with data kept only for t <= tmax.

savefigs(fname=None, onefile=True)[source]#

Saves all current figures to PDF, either to one file or each to a separate file.

Parameters
fname: string, optional

Name of file. If unspecified, the name of the experiment is used.

onefile: boolean, optional

If False, each figures is save to its own PDF file.

savelog(fname=None)[source]#

Save log to file.

Parameters – fname: string, optional

The name of the file. If unspecified, the name of the experiment.

showwells(concise=False, sortby=False, experiments='all', conditions='all', strains='all', experimentincludes=False, experimentexcludes=False, conditionincludes=False, conditionexcludes=False, strainincludes=False, strainexcludes=False)[source]#

Displays wells for specified experiments, conditions, and strains.

Parameters
concise: boolean

If True, display as experiment: condition: strain.

sortby: list of strings, optional

List of column names on which to sort the results.

experiments: string or list of strings

The experiments to include.

conditions: string or list of strings

The conditions to include.

strains: string or list of strings

The strains to include.

experimentincludes: string, optional

Selects only experiments that include the specified string in their name.

experimentexcludes: string, optional

Ignores experiments that include the specified string in their name.

conditionincludes: string, optional

Selects only conditions that include the specified string in their name.

conditionexcludes: string, optional

Ignores conditions that include the specified string in their name.

strainincludes: string, optional

Selects only strains that include the specified string in their name.

strainexcludes: string, optional

Ignores strains that include the specified string in their name.

Examples

>>> p.showwells()
>>> p.showwells(strains= 'Mal12:GFP', conditions= '1% Mal')
webhelp(browser=None)[source]#

Opens detailed examples of how to use in omniplate in a web browser.

Parameters
browser: string, optional

The browser to use - either the default if unspecified or ‘firefox’, ‘chrome’, etc.