notebook.community

Edit and run

Plate reader data analysis

Table of Contents

  1. Global imports, variables and functions
  2. Initiation rates used for Run 3 initiation mutant simulations (for Fig. 4 and Fig. 4--Figure supplement 1)
  3. Single mutant data used for Run 3 initiation mutant simulations (for Fig. 4 and Fig. 4--Figure supplement 1)
  4. Single mutant data used for Run 4 double mutant simulations (for Fig. 5--Figure supplement 1)
  5. Serine single mutant data used for Run 14 initiation rate simulations (for Fig. 4--Figure supplement 1)
  6. Serine single mutant data used for Run 15 double mutant simulations (for Fig. 5--Figure supplement 1)
  7. Single mutant data used for Run 16 multiple CTA mutant simulations (for Fig. 5)
  8. Single mutant data used for Run 5 distance CTC mutant simulations (for Fig. 6 --Figure supplement 1)
  9. Single mutant data used for Run 11 distance CTA mutant simulations (for Fig. 6)
  10. Leucine initiation mutant data used for comparison with Run 3 simulations (for Fig. 4, Fig. 4--Figure supplement 1)
  11. Serine initiation mutant data used for comparison with Run 14 simulations (for Fig. 4--Figure supplement 1)
  12. Leucine multiple mutant data used for comparison with Run 16 simulations (for Fig. 5)
  13. Double mutant data used for comparison with Run 4 simulations (for Fig. 5--Figure supplement 1)
  14. Serine double mutant data used for comparison with Run 15 simulations (for Fig. 5--Figure supplement 1)
  15. CTA distance mutant data used for comparison with Run 11 simulations (for Fig. 6)
  16. CTC distance mutant data used for comparison with Run 5 simulations (for Fig. 6--Figure supplement 1)
  17. Chrom vs. Plasmid comparison to address reviewer's comments (Author Response Fig. 1, 2)

The run numbers above refer to the simulation runs in scripts/run_simulation.ipynb.

Global imports, variables and functions





In [1]:



    


%matplotlib inline
import pandas as pd
from scipy.optimize import curve_fit
import numpy as np
from scipy.stats import tsem
from IPython.display import display, Markdown
import matplotlib.pyplot as plt
import seaborn as sns
sns.set_style('ticks')

od_background = 0.047  # based on well with just growth medium
yfp_background = 69  # based on well with just growth medium

fitpoints = {
    'rich': {
        '2016-01-14': {
            'start': 120,
            'end': 300
        },
        '2016-01-27': {
            'start': 120,
            'end': 300
        },
        '2016-02-04': {
            'start': 120,
            'end': 300
        },
        '2016-03-14': {
            'start': 120,
            'end': 300
        },
        '2016-04-19': {
            'start': 120,
            'end': 300
        },
        '2017-02-01': {
            'start': 120,
            'end': 300
        },
        '2017-02-16': {
            'ser': {
                'start': 120,
                'end': 300
            },
            'leu': {
                'start': 120,
                'end': 300
            },
        },
    },
    'starvation': {
        '2016-01-14': {
            'start': 250,
            'end': 600
        },
        '2016-01-27': {
            'start': 250,
            'end': 600
        },
        '2016-02-04': {
            'start': 250,
            'end': 600
        },
        '2016-03-14': {
            'start': 250,
            'end': 600
        },
        '2016-04-19': {
            'start': 250,
            'end': 600
        },
        '2017-02-01': {
            'start': 250,
            'end': 600
        },
        '2017-02-16': {
            'ser': {
                'start': 200,
                'end': 400
            },
            'leu': {
                'start': 250,
                'end': 450
            },
        },
    }
}


# find points for fitting synthesis rate during rich and starvation
def check_fit(row, growth=['rich', 'starvation']):
    date = row['date']
    time = row['time']
    if date not in fitpoints[growth].keys():
        raise KeyError
    if 'limited_aa' in row:
        aa = row['limited_aa']
        start = fitpoints[growth][date][aa]['start']
        end = fitpoints[growth][date][aa]['end']
    else:
        start = fitpoints[growth][date]['start']
        end = fitpoints[growth][date]['end']
    return (time >= start and time <= end)


# linear function  used for fitting fluorescence data during starvation
def Poly1(x, a, b):
    return b * x / 60 - a / 60


# find synthesis rate for single sample during rich and starvation conditions
def find_synthesis_rate(df):
    # presynthesis rate
    # sort by time and reset index to serial integers
    df = df.sort_values(by='time').reset_index()

    #excludedPoints = y["time"] < pointsForPreFit[y['date']]['start']
    #startingPoint = np.argmax(y['time'][excludedPoints]) + 1
    #minimumPoint = np.argmin(y['yfp/od'][pointsForPreRate]) + startingPoint
    # the average over 3 points where the yfp/od hits minimum is calculated
    #return np.mean(y['yfp/od'][minimumPoint-2:minimumPoint+1])

    # find min pre-starv yfp/od
    minidx = df['yfp/od'][df['richfit']].argmin()
    # find average of yfp/od for 3 points upto min
    richrate = df.loc[minidx - 2:minidx, 'yfp/od'].mean()

    # fit straight line to starvation phase
    BestFitParam, BestFitCOV = curve_fit(Poly1,
                                         list(df['time'][df['starvefit']]),
                                         list(df['yfp'][df['starvefit']]))
    # slope is post rate
    starverate = BestFitParam[1]
    # print(df.drop('time', 1).ix[0])
    return (df.drop([
        'time', 'od', 'yfp', 'richfit', 'starvefit', 'yfp/od', 'index', 'date',
        'plate', 'well'
    ], 1).ix[0].append(
        pd.Series({
            'richrate': richrate,
            'starverate': starverate
        })))


# get synthesis rate for all samples
def get_synthesis_rate(df):
    df['od'] = df['od'] - od_background
    df['yfp'] = df['yfp'] - yfp_background
    df['yfp/od'] = df['yfp'] / df['od']

    # calculate time points for rich and starvation YFP synthesis rates
    df['richfit'] = df.apply(lambda row: check_fit(row, 'rich'), axis=1)
    df['starvefit'] = df.apply(
        lambda row: check_fit(row, 'starvation'), axis=1)

    # group all data points for each well
    wells = df.groupby(['date', 'plate', 'well'])

    # find pre and post synthesis rates for each label
    welldata = wells.apply(find_synthesis_rate).reset_index()
    return welldata


# average synthesis rate over replicate wells
def get_av_synthesis_rate(df):
    # groupby date and sampelname and calculate mean and stderr
    groupedReplicates = df.groupby('sample')
    average = groupedReplicates[['richrate', 'starverate']].agg(np.mean).rename(
        columns={'richrate': 'richrate_mean',
                 'starverate': 'starverate_mean'})
    error = groupedReplicates[['richrate', 'starverate']].agg(tsem).rename(
        columns={'richrate': 'richrate_err',
                 'starverate': 'starverate_err'})
    wellannotations = df.drop(['richrate', 'starverate'],
                              1).groupby('sample').first()
    averagedData = pd.concat(
        [average, error, wellannotations], join='inner', axis=1)
    return averagedData

Initiation rates used for Run 3 initiation mutant simulations (for Fig. 4 and Fig. 4--Figure supplement 1)





In [2]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata1 = pd.read_table('../rawdata/platereader/initiation_cta_data.tsv')
annotations1 = pd.read_table(
    '../rawdata/platereader/initiation_cta_annotations.tsv')
rawdata1 = rawdata1.merge(annotations1, on=['plate', 'well'], how='inner')
rawdata2 = pd.read_table('../rawdata/platereader/initiation_ctc_ctt_data.tsv')
annotations2 = pd.read_table(
    '../rawdata/platereader/initiation_ctc_ctt_annotations.tsv')
rawdata2 = rawdata2.merge(annotations2, on=['plate', 'well'], how='inner')
rawdata = pd.concat([rawdata1, rawdata2])

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)

# group by all variants having the same initiation region
groupedReplicates = welldata[['richrate', 'initiation']].groupby('initiation')

# infer mean initiation rate for each set of initiation rate mutants
# normalize by the 'wt' mutant
inferredInitRate = groupedReplicates.agg(np.mean)
inferredInitRate = inferredInitRate / inferredInitRate['richrate'].ix['wt']
inferredInitRate.rename(
    columns={'richrate': 'inferredInitiationRate'}, inplace=True)
inferredInitRate[['inferredInitiationRate']].to_csv(
    '../processeddata/platereader/inferred_initiationrates_for_initiation_simulations.tsv',
    sep='\t')
inferredInitRate



    


















    
Out[2]:










  
    

      
      inferredInitiationRate
    

    

      initiation
      
    

  
  
    

      ATC
      0.477902
    

    

      RBS2
      0.731301
    

    

      RBS4
      1.112574
    

    

      TTG
      0.337290
    

    

      wt
      1.000000

Single mutant data used for Run 3 initiation mutant simulations (for Fig. 4 and Fig. 4--Figure supplement 1)





In [3]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata1 = pd.read_table('../rawdata/platereader/initiation_cta_data.tsv')
annotations1 = pd.read_table(
    '../rawdata/platereader/initiation_cta_annotations.tsv')
rawdata1 = rawdata1.merge(annotations1, on=['plate', 'well'], how='inner')
rawdata2 = pd.read_table('../rawdata/platereader/initiation_ctc_ctt_data.tsv')
annotations2 = pd.read_table(
    '../rawdata/platereader/initiation_ctc_ctt_annotations.tsv')
rawdata2 = rawdata2.merge(annotations2, on=['plate', 'well'], how='inner')
rawdata = pd.concat([rawdata1, rawdata2])

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# select only wt RBS samples
subsetdata = sampledata[((sampledata['initiation'] == 'wt'))]
# normalize by wt RBS variant without any pause site
normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
normalizationdata = float(subsetdata.loc[normindex, 'starverate_mean'])
# select only single stall site samples
subsetdata = subsetdata[~normindex]
subsetdata = subsetdata[subsetdata['pauselocation'].apply(
    lambda x: len(x.split(',')) == 1)]
# normalization
subsetdata['measuredRateNormalized'] = subsetdata[
    'starverate_mean'] / normalizationdata
# define label
subsetdata['label'] = subsetdata.apply(
    lambda row: row['pausecodon'].lower() + row['pauselocation'], axis=1)
subsetdata['pauselocation'] = subsetdata['pauselocation'].apply(int)
subsetdata = subsetdata.sort_values(by=['pausecodon', 'pauselocation'])
summarydata = subsetdata[['label', 'measuredRateNormalized']]
summarydata = summarydata.set_index('label')
summarydata.to_csv(
    '../processeddata/platereader/measured_yfprates_for_initiation_simulations.tsv',
    sep='\t')
summarydata



    


















    
Out[3]:










  
    

      
      measuredRateNormalized
    

    

      label
      
    

  
  
    

      cta6
      0.303928
    

    

      cta10
      0.229461
    

    

      cta14
      0.385538
    

    

      cta18
      0.344537
    

    

      ctc10
      0.582338
    

    

      ctc14
      0.651636
    

    

      ctt6
      0.423412
    

    

      ctt10
      0.318042

Single mutant data used for Run 4 double mutant simulations (for Fig. 5--Figure supplement 1)





In [4]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/doubles_data.tsv')
annotations = pd.read_table('../rawdata/platereader/doubles_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata.apply(
    lambda row: row['strain_id'][:-2] + '_' + str(row['plate']), axis=1)
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# normalize by max value within each plate (corresponding to no pause variant)
sampledata['measuredRateNormalized'] = sampledata.groupby(
    'plate')['starverate_mean'].transform(lambda x: x / x.max())
subsetdata = sampledata
# select only single stall site samples
normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
subsetdata = subsetdata[~normindex]
subsetdata = subsetdata[subsetdata['pauselocation'].apply(
    lambda x: len(x.split(',')) == 1)]
# define label
subsetdata['label'] = subsetdata.apply(
    lambda row: row['pausecodon'].lower() + row['pauselocation'], axis=1)
subsetdata['pauselocation'] = subsetdata['pauselocation'].apply(int)
subsetdata = subsetdata.sort_values(by=['pausecodon', 'pauselocation'])
summarydata = subsetdata[['label', 'measuredRateNormalized']]
summarydata = summarydata.set_index('label')
summarydata.to_csv(
    '../processeddata/platereader/measured_yfprates_for_double_simulations.tsv',
    sep='\t')
summarydata



    


















    
Out[4]:










  
    

      
      measuredRateNormalized
    

    

      label
      
    

  
  
    

      cta2
      0.171785
    

    

      cta6
      0.224788
    

    

      cta10
      0.158591
    

    

      cta14
      0.316839
    

    

      cta18
      0.297784
    

    

      ctc2
      0.407218
    

    

      ctc6
      0.608872
    

    

      ctc10
      0.747329
    

    

      ctc14
      0.818875
    

    

      ctc18
      0.813529
    

    

      ctt2
      0.373273
    

    

      ctt6
      0.525894
    

    

      ctt10
      0.431487
    

    

      ctt14
      0.789663
    

    

      ctt18
      0.733365

Serine single mutant data used for Run 14 initiation rate simulations (for Fig. 4--Figure supplement 1)





In [5]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/serine_multiple_cta_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/serine_multiple_cta_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()
# select only wt RBS samples
subsetdata = sampledata[((sampledata['initiation_variant'] == 'wt'))]
# normalize by wt RBS variant without any pause site
normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
normalizationdata = float(subsetdata.loc[normindex, 'starverate_mean'])
# select only single stall samples
subsetdata = subsetdata[~normindex]
subsetdata = subsetdata[subsetdata['pauselocation'].apply(lambda x: x == '5')]
# normalization
subsetdata['measuredRateNormalized'] = subsetdata[
    'starverate_mean'] / normalizationdata
# define label
subsetdata['label'] = subsetdata.apply(
    lambda row: row['pausecodon'].lower() + row['pauselocation'], axis=1)
subsetdata['pauselocation'] = subsetdata['pauselocation'].apply(int)
subsetdata = subsetdata.sort_values(by=['pausecodon', 'pauselocation'])
summarydata = subsetdata[['label', 'measuredRateNormalized']]
summarydata = summarydata.set_index('label')
summarydata.to_csv(
    '../processeddata/platereader/measured_yfprates_for_serine_initiation_simulations.tsv',
    sep='\t',
    index_label='')
summarydata



    


















    
Out[5]:










  
    

      
      measuredRateNormalized
    

    

      label
      
    

  
  
    

      tcg5
      0.809052

Serine single mutant data used for Run 15 double mutant simulations (for Fig. 5--Figure supplement 1)





In [6]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/serine_multiple_cta_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/serine_multiple_cta_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# select only serine mutants used for double mutant expt
sampledata = sampledata[sampledata['initiation_variant'].apply(
    lambda x: pd.isnull(x) or x == 'wt')]
sampledata = sampledata[sampledata['limited_aa'].apply(lambda x: x == 'ser')]

# normalize by max value within each plate (corresponding to no pause variant)
sampledata['measuredRateNormalized'] = sampledata.groupby(
    'plate')['starverate_mean'].transform(lambda x: x / x.max())
subsetdata = sampledata
# select only single stall site samples
normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
subsetdata = subsetdata[~normindex]
subsetdata = subsetdata[subsetdata['pauselocation'].apply(
    lambda x: len(x.split(',')) == 1)]
# define label
subsetdata['label'] = subsetdata.apply(
    lambda row: row['pausecodon'].lower() + row['pauselocation'], axis=1)
subsetdata['pauselocation'] = subsetdata['pauselocation'].apply(int)
subsetdata = subsetdata.sort_values(by=['pausecodon', 'pauselocation'])
summarydata = subsetdata[['label', 'measuredRateNormalized']]
summarydata = summarydata.set_index('label')

summarydata.to_csv(
    '../processeddata/platereader/measured_yfprates_for_serine_double_simulations.tsv',
    sep='\t',
    index_label='')
summarydata



    


















    
Out[6]:










  
    

      
      measuredRateNormalized
    

    

      label
      
    

  
  
    

      tcg2
      0.676833
    

    

      tcg3
      0.683100
    

    

      tcg4
      0.789628
    

    

      tcg5
      0.809052
    

    

      tcg6
      0.740441
    

    

      tcg7
      0.844806
    

    

      tcg8
      0.890436

Single mutant data used for Run 16 multiple CTA mutant simulations (for Fig. 5)





In [7]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/serine_multiple_cta_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/serine_multiple_cta_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# select only serine mutants used for double mutant expt
sampledata = sampledata[sampledata['limited_aa'].apply(lambda x: x == 'leu')]

# normalize by max value within each plate (corresponding to no pause variant)
sampledata['measuredRateNormalized'] = sampledata.groupby(
    'plate')['starverate_mean'].transform(lambda x: x / x.max())
subsetdata = sampledata
# select only single stall site samples
normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
subsetdata = subsetdata[~normindex]
subsetdata = subsetdata[subsetdata['pauselocation'].apply(
    lambda x: len(x.split(',')) == 1)]
# define label
subsetdata['label'] = subsetdata.apply(
    lambda row: row['pausecodon'].lower() + row['pauselocation'], axis=1)
subsetdata['pauselocation'] = subsetdata['pauselocation'].apply(int)
subsetdata = subsetdata.sort_values(by=['pausecodon', 'pauselocation'])
summarydata = subsetdata[['label', 'measuredRateNormalized']]
summarydata = summarydata.set_index('label')

summarydata.to_csv(
    '../processeddata/platereader/measured_yfprates_for_leucine_multiple_simulations.tsv',
    sep='\t',
    index_label='')
summarydata



    


















    
Out[7]:










  
    

      
      measuredRateNormalized
    

    

      label
      
    

  
  
    

      cta2
      0.190919
    

    

      cta6
      0.272954
    

    

      cta10
      0.204786
    

    

      cta14
      0.333250
    

    

      cta18
      0.286355

Single mutant data used for Run 5 distance CTC mutant simulations (for Fig. 6 --Figure supplement 1)





In [8]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/distance_ctc_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/distance_ctc_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# normalize by max value (corresponding to no pause variant)
sampledata['measuredRateNormalized'] = sampledata[
    'starverate_mean'] / sampledata['starverate_mean'].max()
subsetdata = sampledata
# select only single stall site samples
normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
subsetdata = subsetdata[~normindex]
subsetdata = subsetdata[subsetdata['pauselocation'].apply(
    lambda x: len(x.split(',')) == 1)]
# define label
subsetdata['label'] = subsetdata.apply(
    lambda row: row['pausecodon'].lower() + row['pauselocation'], axis=1)
subsetdata['pauselocation'] = subsetdata['pauselocation'].apply(int)
subsetdata = subsetdata.sort_values(by=['pausecodon', 'pauselocation'])
summarydata = subsetdata[['label', 'measuredRateNormalized']]
summarydata = summarydata.set_index('label')

summarydata.to_csv(
    '../processeddata/platereader/measured_yfprates_for_distance_simulations.tsv',
    sep='\t')
summarydata



    


















    
Out[8]:










  
    

      
      measuredRateNormalized
    

    

      label
      
    

  
  
    

      ctc8
      0.687575
    

    

      ctc9
      0.646659
    

    

      ctc10
      0.681533
    

    

      ctc11
      0.618805
    

    

      ctc12
      0.622357
    

    

      ctc13
      0.713587
    

    

      ctc14
      0.796001

Single mutant data used for Run 11 distance CTA mutant simulations (for Fig. 6)





In [9]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/distance_cta_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/distance_cta_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# normalize by max value (corresponding to no pause variant)
sampledata['measuredRateNormalized'] = sampledata[
    'starverate_mean'] / sampledata['starverate_mean'].max()
subsetdata = sampledata
# select only single stall site samples
normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
subsetdata = subsetdata[~normindex]
subsetdata = subsetdata[subsetdata['pauselocation'].apply(
    lambda x: len(x.split(',')) == 1)]
# define label
subsetdata['label'] = subsetdata.apply(
    lambda row: row['pausecodon'].lower() + row['pauselocation'], axis=1)
subsetdata['pauselocation'] = subsetdata['pauselocation'].apply(int)
subsetdata = subsetdata.sort_values(by=['pausecodon', 'pauselocation'])
summarydata = subsetdata[['label', 'measuredRateNormalized']]
summarydata = summarydata.set_index('label')

summarydata.to_csv(
    '../processeddata/platereader/measured_yfprates_for_cta_distance_simulations.tsv',
    sep='\t')
summarydata



    


















    
Out[9]:










  
    

      
      measuredRateNormalized
    

    

      label
      
    

  
  
    

      cta8
      0.325880
    

    

      cta9
      0.227703
    

    

      cta10
      0.206188
    

    

      cta11
      0.270118
    

    

      cta12
      0.285635
    

    

      cta13
      0.284764
    

    

      cta14
      0.314228

Leucine initiation mutant data used for comparison with Run 3 simulations (for Fig. 4, Fig. 4--Figure supplement 1)





In [2]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata1 = pd.read_table('../rawdata/platereader/initiation_cta_data.tsv')
annotations1 = pd.read_table(
    '../rawdata/platereader/initiation_cta_annotations.tsv')
rawdata1 = rawdata1.merge(annotations1, on=['plate', 'well'], how='inner')
rawdata2 = pd.read_table('../rawdata/platereader/initiation_ctc_ctt_data.tsv')
annotations2 = pd.read_table(
    '../rawdata/platereader/initiation_ctc_ctt_annotations.tsv')
rawdata2 = rawdata2.merge(annotations2, on=['plate', 'well'], how='inner')
rawdata = pd.concat([rawdata1, rawdata2])

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

subsetdata = sampledata[((sampledata['initiation'].apply(
    lambda x: not pd.isnull(x))))]

normalization = subsetdata.loc[subsetdata['pausecodon'].apply(
    lambda x: pd.isnull(x)
), ['starverate_mean', 'starverate_err', 'initiation']].set_index('initiation')

for row in subsetdata[['pausecodon', 'pauselocation']].drop_duplicates(
        keep='first').iterrows():
    pausecodon = row[1]['pausecodon']
    pauselocation = row[1]['pauselocation']
    if pd.isnull(pausecodon):  # exclude no pause control
        continue
    if len(pauselocation.split(',')) != 1:  # exclude many pause control
        continue
    summarydata = subsetdata.loc[
        ((subsetdata['pauselocation'] == pauselocation) &
         (subsetdata['pausecodon'] == pausecodon)), [
             'starverate_mean', 'starverate_err', 'initiation', 'richrate_mean'
         ]].set_index('initiation')

    normalizationdata = normalization.ix[summarydata.index]

    summarydata.loc[:, 'starverate_err'] = (
        summarydata.loc[:, 'starverate_err'].values /
        summarydata.loc[:, 'starverate_mean'].values +
        normalizationdata.loc[:, 'starverate_err'].values /
        normalizationdata.loc[:, 'starverate_mean'].values)

    summarydata.loc[:, 'starverate_mean'] = (
        summarydata.loc[:, 'starverate_mean'] /
        normalizationdata.loc[:, 'starverate_mean'])

    summarydata.loc[:,
                    'starverate_err'] *= summarydata.loc[:,
                                                         'starverate_mean'].values

    summarydata.to_csv(
        '../processeddata/platereader/measured_yfprates_' + pausecodon.lower()
        + pauselocation + '_initiation_mutants.tsv',
        sep='\t')
    display(Markdown('**{}{}**'.format(pausecodon, pauselocation)))
    display(summarydata)



    


















    







CTT10











    










  
    

      
      starverate_mean
      starverate_err
      richrate_mean
    

    

      initiation
      
      
      
    

  
  
    

      wt
      0.318042
      0.017261
      38689.654350
    

    

      TTG
      0.648113
      0.052457
      12562.142331
    

    

      ATC
      0.511094
      0.026543
      20642.153810
    

    

      RBS2
      0.472924
      0.013474
      30853.389552
    

    

      RBS4
      0.341356
      0.018972
      47207.684935
    

  













    







CTC10











    










  
    

      
      starverate_mean
      starverate_err
      richrate_mean
    

    

      initiation
      
      
      
    

  
  
    

      TTG
      0.840797
      0.041461
      12355.518308
    

    

      ATC
      0.763430
      0.023606
      20036.156523
    

    

      RBS2
      0.709053
      0.032872
      29995.923527
    

    

      RBS4
      0.621765
      0.028562
      47036.552791
    

    

      wt
      0.582338
      0.023900
      43956.361634
    

  













    







CTC14











    










  
    

      
      starverate_mean
      starverate_err
      richrate_mean
    

    

      initiation
      
      
      
    

  
  
    

      TTG
      0.889600
      0.070675
      11902.140816
    

    

      ATC
      0.821194
      0.031650
      18746.807511
    

    

      RBS2
      0.884476
      0.073296
      33438.206385
    

    

      RBS4
      0.715794
      0.030441
      46088.268097
    

    

      wt
      0.651636
      0.030557
      40943.525748
    

  













    







CTT6











    










  
    

      
      starverate_mean
      starverate_err
      richrate_mean
    

    

      initiation
      
      
      
    

  
  
    

      TTG
      0.678632
      0.062761
      13146.031332
    

    

      ATC
      0.543485
      0.023146
      21421.899946
    

    

      RBS2
      0.472156
      0.055165
      30642.361761
    

    

      RBS4
      0.385909
      0.009965
      47100.740977
    

    

      wt
      0.423412
      0.026079
      46098.868902
    

  













    







CTA6











    










  
    

      
      starverate_mean
      starverate_err
      richrate_mean
    

    

      initiation
      
      
      
    

  
  
    

      TTG
      0.443839
      0.052841
      16269.396907
    

    

      RBS2
      0.332863
      0.020336
      32849.301552
    

    

      RBS4
      0.251209
      0.029883
      47062.903358
    

    

      wt
      0.303928
      0.022081
      44343.058633
    

  













    







CTA10











    










  
    

      
      starverate_mean
      starverate_err
      richrate_mean
    

    

      initiation
      
      
      
    

  
  
    

      TTG
      0.368502
      0.041456
      15840.469128
    

    

      ATC
      0.292160
      0.018617
      23296.265919
    

    

      RBS2
      0.280564
      0.030081
      31238.267866
    

    

      RBS4
      0.220833
      0.005356
      47203.079344
    

    

      wt
      0.229461
      0.017871
      39272.351020
    

  













    







CTA14











    










  
    

      
      starverate_mean
      starverate_err
      richrate_mean
    

    

      initiation
      
      
      
    

  
  
    

      TTG
      0.503191
      0.038305
      15686.145812
    

    

      ATC
      0.473015
      0.024920
      20681.257780
    

    

      RBS2
      0.445447
      0.020330
      31292.032778
    

    

      RBS4
      0.318552
      0.013651
      48336.911754
    

    

      wt
      0.385538
      0.016694
      41167.261358
    

  













    







CTA18











    










  
    

      
      starverate_mean
      starverate_err
      richrate_mean
    

    

      initiation
      
      
      
    

  
  
    

      TTG
      0.621896
      0.048745
      14358.944185
    

    

      ATC
      0.504772
      0.025518
      18994.227312
    

    

      RBS2
      0.412495
      0.023299
      29720.540493
    

    

      RBS4
      0.318594
      0.022771
      47802.319668
    

    

      wt
      0.344537
      0.025993
      43414.549043

Serine initiation mutant data used for comparison with Run 14 simulations (for Fig. 4--Figure supplement 1)





In [11]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/serine_multiple_cta_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/serine_multiple_cta_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

subsetdata = sampledata[((sampledata['initiation_variant'].apply(
    lambda x: not pd.isnull(x))) & (sampledata['limited_aa'].apply(
        lambda x: not pd.isnull(x))))]

normalizationdata = subsetdata.loc[
    subsetdata['pausecodon'].apply(lambda x: pd.isnull(x)),
    ['starverate_mean', 'starverate_err', 'initiation_variant']].set_index(
        'initiation_variant')

pausecodons = subsetdata['pausecodon'].unique()
pausecodon = pausecodons[0]

for pauselocation in subsetdata['pauselocation'].unique():
    if pauselocation != '5':
        continue
    summarydata = subsetdata.loc[
        subsetdata['pauselocation'] == pauselocation,
        ['starverate_mean', 'starverate_err', 'initiation_variant']].set_index(
            'initiation_variant')
    summarydata.loc[:, 'starverate_err'] = (
        summarydata.loc[:, 'starverate_err'].values /
        summarydata.loc[:, 'starverate_mean'].values +
        normalizationdata.loc[:, 'starverate_err'].values /
        normalizationdata.loc[:, 'starverate_mean'].values)

    summarydata.loc[:, 'starverate_mean'] = (
        summarydata.loc[:, 'starverate_mean'] /
        normalizationdata.loc[:, 'starverate_mean'])

    summarydata.loc[:,
                    'starverate_err'] *= summarydata.loc[:,
                                                         'starverate_mean'].values

    summarydata.to_csv(
        '../processeddata/platereader/measured_yfprates_' + pausecodon.lower()
        + pauselocation + '_initiation_mutants.tsv',
        sep='\t')
    display(Markdown('**{}{}**'.format(pausecodon, pauselocation)))
    display(summarydata)



    


















    







TCG5











    










  
    

      
      starverate_mean
      starverate_err
    

    

      initiation_variant
      
      
    

  
  
    

      wt
      0.809052
      0.023178
    

    

      TTG
      1.098876
      0.037491
    

    

      ATC
      0.913414
      0.111765
    

    

      RBS2
      0.890407
      0.019397
    

    

      RBS4
      0.821327
      0.061075

Leucine multiple mutant data used for comparison with Run 16 simulations (for Fig. 5)





In [12]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/serine_multiple_cta_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/serine_multiple_cta_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# select only serine mutants used for double mutant expt
sampledata = sampledata[sampledata['limited_aa'].apply(lambda x: x == 'leu')]

# normalize by max value within each plate (corresponding to no pause variant)
norm = sampledata['starverate_mean'].max()
sampledata['starverate_err'] = sampledata['starverate_err'] / norm
sampledata['starverate_mean'] = sampledata['starverate_mean'] / norm
subsetdata = sampledata
# select only 1-4 stall samples
normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
subsetdata = subsetdata[~normindex]
subsetdata = subsetdata[~subsetdata['pauselocation'].apply(
    lambda x: len(x.split(',')) > 4)]
# define label
summarydata = subsetdata[
    ['pausecodon', 'pauselocation', 'starverate_mean', 'starverate_err']]
summarydata.to_csv(
    '../processeddata/platereader/measured_yfprates_cta_multiple_mutants.tsv',
    sep='\t',
    index=False)
summarydata



    


















    
Out[12]:










  
    

      
      pausecodon
      pauselocation
      starverate_mean
      starverate_err
    

  
  
    

      35
      CTA
      6,10,14
      0.108193
      0.023922
    

    

      36
      CTA
      6,10,18
      0.105813
      0.031174
    

    

      37
      CTA
      6,10,14,18
      0.074351
      0.035449
    

    

      38
      CTA
      2
      0.190919
      0.029384
    

    

      39
      CTA
      6
      0.272954
      0.007965
    

    

      40
      CTA
      10
      0.204786
      0.030056
    

    

      41
      CTA
      14
      0.333250
      0.040942
    

    

      42
      CTA
      18
      0.286355
      0.005394
    

    

      44
      CTA
      2,14
      0.124962
      0.034798
    

    

      45
      CTA
      2,18
      0.166188
      0.045410
    

    

      46
      CTA
      6,14
      0.111677
      0.026926
    

    

      47
      CTA
      6,18
      0.169649
      0.014521
    

    

      48
      CTA
      10,14
      0.160739
      0.029566
    

    

      49
      CTA
      10,18
      0.266150
      0.014324
    

    

      50
      CTA
      2,6
      0.071238
      0.044979
    

    

      51
      CTA
      14,18
      0.191745
      0.023999
    

    

      52
      CTA
      2,10
      0.066953
      0.033681
    

    

      53
      CTA
      6,10
      0.145890
      0.041536

Double mutant data used for comparison with Run 4 simulations (for Fig. 5--Figure supplement 1)





In [13]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/doubles_data.tsv')
annotations = pd.read_table('../rawdata/platereader/doubles_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata.apply(
    lambda row: row['strain_id'][:-2] + '_' + str(row['plate']), axis=1)
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# normalize by max value within each plate (corresponding to no pause variant)
norm = sampledata.groupby('plate')['starverate_mean'].transform(
    lambda x: x.max())
sampledata['starverate_err'] = sampledata['starverate_err'] / norm
sampledata['starverate_mean'] = sampledata['starverate_mean'] / norm
for pausecodon in sampledata['pausecodon'].unique():
    if pausecodon not in ['CTC', 'CTT', 'CTA']:
        continue
    subsetdata = sampledata
    # select only single stall site samples
    subsetdata = subsetdata[subsetdata['pausecodon'] == pausecodon]
    normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
    subsetdata = subsetdata[~normindex]
    subsetdata = subsetdata[~subsetdata['pauselocation'].apply(
        lambda x: len(x.split(',')) > 4)]
    summarydata = subsetdata[
        ['pausecodon', 'pauselocation', 'starverate_mean', 'starverate_err']]
    summarydata.to_csv(
        '../processeddata/platereader/measured_yfprates_{}_double_mutants.tsv'.
        format(pausecodon.lower()),
        sep='\t',
        index=False)
    display(Markdown('**{}**'.format(pausecodon)))
    display(summarydata)



    


















    







CTT











    










  
    

      
      pausecodon
      pauselocation
      starverate_mean
      starverate_err
    

  
  
    

      0
      CTT
      10
      0.431487
      0.010396
    

    

      1
      CTT
      14
      0.789663
      0.028166
    

    

      2
      CTT
      18
      0.733365
      0.007182
    

    

      16
      CTT
      2,14
      0.362381
      0.008304
    

    

      17
      CTT
      2,18
      0.392638
      0.006299
    

    

      18
      CTT
      6,14
      0.457245
      0.007891
    

    

      19
      CTT
      6,18
      0.440528
      0.028604
    

    

      20
      CTT
      10,14
      0.428675
      0.026286
    

    

      21
      CTT
      10,18
      0.395161
      0.021789
    

    

      22
      CTT
      2,6
      0.251769
      0.009749
    

    

      23
      CTT
      14,18
      0.616904
      0.018653
    

    

      24
      CTT
      2,10
      0.218175
      0.013362
    

    

      25
      CTT
      6,10
      0.263784
      0.020677
    

    

      49
      CTT
      2
      0.373273
      0.014996
    

    

      50
      CTT
      6
      0.525894
      0.018002
    

  













    







CTA











    










  
    

      
      pausecodon
      pauselocation
      starverate_mean
      starverate_err
    

  
  
    

      26
      CTA
      2
      0.171785
      0.021595
    

    

      27
      CTA
      6
      0.224788
      0.019202
    

    

      28
      CTA
      10
      0.158591
      0.016815
    

    

      29
      CTA
      14
      0.316839
      0.005631
    

    

      30
      CTA
      18
      0.297784
      0.011593
    

    

      34
      CTA
      2,14
      0.102935
      0.010689
    

    

      35
      CTA
      2,18
      0.170940
      0.003687
    

    

      36
      CTA
      6,14
      0.146685
      0.007884
    

    

      37
      CTA
      6,18
      0.156445
      0.017380
    

    

      38
      CTA
      10,14
      0.108254
      0.005452
    

    

      39
      CTA
      10,18
      0.151476
      0.015021
    

    

      40
      CTA
      2,6
      0.093039
      0.031273
    

    

      41
      CTA
      14,18
      0.160114
      0.009698
    

    

      42
      CTA
      2,10
      0.076930
      0.012757
    

    

      43
      CTA
      6,10
      0.070348
      0.008905
    

  













    







CTC











    










  
    

      
      pausecodon
      pauselocation
      starverate_mean
      starverate_err
    

  
  
    

      6
      CTC
      2,14
      0.365231
      0.008754
    

    

      7
      CTC
      2,18
      0.428413
      0.008683
    

    

      8
      CTC
      6,14
      0.553870
      0.015503
    

    

      9
      CTC
      6,18
      0.607476
      0.016925
    

    

      10
      CTC
      10,14
      0.642563
      0.020819
    

    

      11
      CTC
      10,18
      0.679807
      0.007946
    

    

      12
      CTC
      2,6
      0.269103
      0.020010
    

    

      13
      CTC
      14,18
      0.712054
      0.013971
    

    

      14
      CTC
      2,10
      0.320510
      0.012594
    

    

      15
      CTC
      6,10
      0.483822
      0.008361
    

    

      44
      CTC
      2
      0.407218
      0.007115
    

    

      45
      CTC
      6
      0.608872
      0.016441
    

    

      46
      CTC
      10
      0.747329
      0.005557
    

    

      47
      CTC
      14
      0.818875
      0.033167
    

    

      48
      CTC
      18
      0.813529
      0.010263

Serine double mutant data used for comparison with Run 15 simulations (for Fig. 5--Figure supplement 1)





In [6]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/serine_multiple_cta_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/serine_multiple_cta_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# select only serine mutants used for double mutant expt
sampledata = sampledata[sampledata['initiation_variant'].apply(
    lambda x: pd.isnull(x) or x == 'wt')]
sampledata = sampledata[sampledata['limited_aa'].apply(lambda x: x == 'ser')]

# normalize by max value within each plate (corresponding to no pause variant)
norm = sampledata['starverate_mean'].max()
sampledata['starverate_err'] = sampledata['starverate_err'] / norm
sampledata['starverate_mean'] = sampledata['starverate_mean'] / norm
for pausecodon in sampledata['pausecodon'].unique():
    if pausecodon not in ['TCG']:
        continue
    subsetdata = sampledata
    # select only single stall site samples
    subsetdata = subsetdata[subsetdata['pausecodon'] == pausecodon]
    normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
    subsetdata = subsetdata[~normindex]
    subsetdata = subsetdata[~subsetdata['pauselocation'].apply(
        lambda x: len(x.split(',')) > 4)]
    # TCG8 was not used for double mutant expt
    subsetdata = subsetdata[subsetdata['pauselocation'].apply(
        lambda x: x != '8')]
    summarydata = subsetdata[
        ['pausecodon', 'pauselocation', 'starverate_mean', 'starverate_err']]
    summarydata.to_csv(
        '../processeddata/platereader/measured_yfprates_{}_double_mutants.tsv'.
        format(pausecodon.lower()),
        sep='\t',
        index=False)
    display(Markdown('**{}**'.format(pausecodon)))
    display(summarydata)



    


















    







TCG











    










  
    

      
      pausecodon
      pauselocation
      starverate_mean
      starverate_err
    

  
  
    

      1
      TCG
      2
      0.676833
      0.010117
    

    

      2
      TCG
      3
      0.683100
      0.011257
    

    

      3
      TCG
      4
      0.789628
      0.006436
    

    

      4
      TCG
      5
      0.809052
      0.013044
    

    

      5
      TCG
      6
      0.740441
      0.022510
    

    

      6
      TCG
      7
      0.844806
      0.006194
    

    

      10
      TCG
      2,5
      0.639850
      0.026359
    

    

      11
      TCG
      2,6
      0.575802
      0.003365
    

    

      12
      TCG
      2,7
      0.620216
      0.007561
    

    

      13
      TCG
      3,5
      0.693729
      0.008726
    

    

      14
      TCG
      3,6
      0.580759
      0.011743
    

    

      15
      TCG
      3,7
      0.628790
      0.007846
    

    

      16
      TCG
      4,5
      0.730769
      0.004271
    

    

      17
      TCG
      4,6
      0.701005
      0.022835
    

    

      18
      TCG
      4,7
      0.735890
      0.007294

CTA distance mutant data used for comparison with Run 11 simulations (for Fig. 6)





In [15]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/distance_cta_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/distance_cta_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# normalize by max value within each plate (corresponding to no pause variant)
norm = sampledata['starverate_mean'].max()
sampledata['starverate_err'] = sampledata['starverate_err'] / norm
sampledata['starverate_mean'] = sampledata['starverate_mean'] / norm
for pausecodon in sampledata['pausecodon'].unique():
    if pausecodon not in ['CTA']:
        continue
    subsetdata = sampledata
    # select only single stall site samples
    subsetdata = subsetdata[subsetdata['pausecodon'] == pausecodon]
    normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
    subsetdata = subsetdata[~normindex]
    subsetdata = subsetdata[~subsetdata['pauselocation'].apply(
        lambda x: len(x.split(',')) > 4)]
    subsetdata = subsetdata[~subsetdata['pauselocation'].apply(
        lambda x: '9,' in x)]
    summarydata = subsetdata[
        ['pausecodon', 'pauselocation', 'starverate_mean', 'starverate_err']]
    summarydata.to_csv(
        '../processeddata/platereader/measured_yfprates_{}_distance_mutants.tsv'.
        format(pausecodon.lower()),
        sep='\t',
        index=False)
    display(Markdown('**{}**'.format(pausecodon)))
    display(summarydata)



    


















    







CTA











    










  
    

      
      pausecodon
      pauselocation
      starverate_mean
      starverate_err
    

  
  
    

      1
      CTA
      8,9
      0.136273
      0.012459
    

    

      2
      CTA
      8,10
      0.101509
      0.023147
    

    

      3
      CTA
      8,11
      0.178172
      0.011986
    

    

      4
      CTA
      8,12
      0.173283
      0.010144
    

    

      5
      CTA
      8,13
      0.156874
      0.002760
    

    

      6
      CTA
      8,14
      0.176990
      0.027423
    

    

      12
      CTA
      11,12
      0.135461
      0.014066
    

    

      13
      CTA
      13,14
      0.123600
      0.008373
    

    

      14
      CTA
      9
      0.227703
      0.014838
    

    

      15
      CTA
      10
      0.206188
      0.006772
    

    

      16
      CTA
      11
      0.270118
      0.005881
    

    

      17
      CTA
      12
      0.285635
      0.016765
    

    

      18
      CTA
      13
      0.284764
      0.008610
    

    

      19
      CTA
      14
      0.314228
      0.004354
    

    

      20
      CTA
      8
      0.325880
      0.005299

CTC distance mutant data used for comparison with Run 5 simulations (for Fig. 6--Figure supplement 1)





In [16]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/distance_ctc_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/distance_ctc_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')

# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()

# normalize by max value within each plate (corresponding to no pause variant)
norm = sampledata['starverate_mean'].max()
sampledata['starverate_err'] = sampledata['starverate_err'] / norm
sampledata['starverate_mean'] = sampledata['starverate_mean'] / norm
for pausecodon in sampledata['pausecodon'].unique():
    if pausecodon not in ['CTC']:
        continue
    subsetdata = sampledata
    # select only single stall site samples
    subsetdata = subsetdata[subsetdata['pausecodon'] == pausecodon]
    normindex = subsetdata['pausecodon'].apply(lambda x: pd.isnull(x))
    subsetdata = subsetdata[~normindex]
    subsetdata = subsetdata[~subsetdata['pauselocation'].apply(
        lambda x: len(x.split(',')) > 4)]
    subsetdata = subsetdata[~subsetdata['pauselocation'].apply(
        lambda x: '13' in x or '14' in x)]
    summarydata = subsetdata[
        ['pausecodon', 'pauselocation', 'starverate_mean', 'starverate_err']]
    summarydata.to_csv(
        '../processeddata/platereader/measured_yfprates_{}_distance_mutants.tsv'.
        format(pausecodon.lower()),
        sep='\t',
        index=False)
    display(Markdown('**{}**'.format(pausecodon)))
    display(summarydata)



    


















    







CTC











    










  
    

      
      pausecodon
      pauselocation
      starverate_mean
      starverate_err
    

  
  
    

      2
      CTC
      8
      0.687575
      0.006808
    

    

      3
      CTC
      9
      0.646659
      0.015281
    

    

      4
      CTC
      8,9
      0.387758
      0.004063
    

    

      5
      CTC
      8,10
      0.410743
      0.010952
    

    

      6
      CTC
      9,10
      0.392448
      0.006810
    

    

      9
      CTC
      11
      0.618805
      0.022442
    

    

      10
      CTC
      12
      0.622357
      0.041011
    

    

      11
      CTC
      8,11
      0.475482
      0.009059
    

    

      12
      CTC
      8,12
      0.514034
      0.005433
    

    

      13
      CTC
      11,12
      0.403571
      0.005961
    

    

      17
      CTC
      9,11
      0.437395
      0.013760
    

    

      18
      CTC
      9,12
      0.475304
      0.004471
    

    

      21
      CTC
      10
      0.681533
      0.000177
    

  




















In [87]:



    


# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/chrom_vs_plasmid_data.tsv')
annotations = pd.read_table(
    '../rawdata/platereader/chrom_vs_plasmid_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')
# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata['strain_id'].apply(lambda x: x[:-2])
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()
chromdata = sampledata[sampledata['gene_location'] == 'chromosome']
#normalize
norm = float(chromdata['starverate_mean'].max())
chromdata['starverate_mean'] = chromdata['starverate_mean'] / norm
chromdata['starverate_err'] = chromdata['starverate_err'] / norm

# read in raw data for each well and corresponding annotations and merge
rawdata = pd.read_table('../rawdata/platereader/doubles_data.tsv')
annotations = pd.read_table('../rawdata/platereader/doubles_annotations.tsv')
rawdata = rawdata.merge(annotations, on=['plate', 'well'], how='inner')
# calculate the yfp synthesis rate for each well
welldata = get_synthesis_rate(rawdata)
# define sample column for averaging over replicates
# this is specific for each experiment
welldata['sample'] = welldata.apply(
    lambda row: row['strain_id'][:-2] + '_' + str(row['plate']), axis=1)
# average synthesis rate over replicates
sampledata = get_av_synthesis_rate(welldata).reset_index()
plasmiddata = sampledata[sampledata['pausecodon'] == 'CTA']
plasmiddata = sampledata[sampledata['plate'] == 1]
plasmiddata = plasmiddata.merge(
    chromdata[['pauselocation']], how='right', on='pauselocation')
#normalize
norm = float(plasmiddata['starverate_mean'].max())
plasmiddata['starverate_mean'] = plasmiddata['starverate_mean'] / norm
plasmiddata['starverate_err'] = plasmiddata['starverate_err'] / norm

# normalize rich for easy comparison
norm = float(plasmiddata['richrate_mean'].max())
chromdata['richrate_mean'] = chromdata['richrate_mean'] / norm
chromdata['richrate_err'] = chromdata['richrate_err'] / norm
plasmiddata['richrate_mean'] = plasmiddata['richrate_mean'] / norm
plasmiddata['richrate_err'] = plasmiddata['richrate_err'] / norm

plasmiddata = plasmiddata.set_index('pauselocation')
chromdata = chromdata.set_index('pauselocation')
plasmiddata = plasmiddata.ix[chromdata.index]



    


















    






/fh/fast/subramaniam_a/user/rasi/virtualenv/default2/lib/python2.7/site-packages/ipykernel/__main__.py:15: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy
/fh/fast/subramaniam_a/user/rasi/virtualenv/default2/lib/python2.7/site-packages/ipykernel/__main__.py:16: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy

Chrom vs. Plasmid comparison to address reviewer's comments (Author Response Fig. 1, 2)





In [88]:



    


fig = plt.figure()
fig.set_size_inches([3, 2])

ax = fig.add_subplot(1, 1, 1)
sns.despine(offset=5)
subset = plasmiddata
ax.errorbar(
    np.arange(len(subset)) - 0.1,
    subset['richrate_mean'],
    subset['richrate_err'],
    color='black',
    capsize=10,
    elinewidth=1,
    linestyle='None',
    marker='^',
    markersize=4,
    label='plasmid')
subset = chromdata
ax.errorbar(
    np.arange(len(subset)) + 0.1,
    subset['richrate_mean'],
    subset['richrate_err'],
    color='green',
    capsize=10,
    elinewidth=1,
    linestyle='None',
    marker='o',
    markersize=4,
    label='chromosome')
xticklabels = map(
    lambda x: str(x).replace(',', ' + ').replace('nan', 'no-stall'),
    subset.index)
xticklabels[1] = '7-stalls'
ax.set_ylim(bottom=-0.1, top=1.1)
ax.set_xticks(np.arange(len(subset)))
ax.set_xticklabels(xticklabels, rotation=45, ha='right')
ax.set(xlabel='Location of stall sites',
       ylabel='YFP synthesis rate (a.u.)\n(Leu rich)')
ax.legend(loc=2, bbox_to_anchor=(1.2, 1))
fig.savefig('../figures/reviewer_response_fig1.svg')



    


















    
























In [90]:



    


fig = plt.figure()
fig.set_size_inches([3, 2])

ax = fig.add_subplot(1, 1, 1)
sns.despine(offset=5)
subset = plasmiddata
ax.errorbar(
    np.arange(len(subset)) - 0.1,
    subset['starverate_mean'],
    subset['starverate_err'],
    color='black',
    capsize=10,
    elinewidth=1,
    linestyle='None',
    marker='^',
    markersize=4,
    label='plasmid')
subset = chromdata
ax.errorbar(
    np.arange(len(subset)) + 0.1,
    subset['starverate_mean'],
    subset['starverate_err'],
    color='green',
    capsize=10,
    elinewidth=1,
    linestyle='None',
    marker='o',
    markersize=4,
    label='chromosome')
xticklabels = map(
    lambda x: str(x).replace(',', ' + ').replace('nan', 'no-stall'),
    subset.index)
xticklabels[1] = '7-stalls'
ax.set_ylim(bottom=-0.1, top=1.1)
ax.set_xticks(np.arange(len(subset)))
ax.set_xticklabels(xticklabels, rotation=45, ha='right')
ax.set(xlabel='Location of stall sites',
       ylabel='YFP synthesis rate (w.r.t no-stall)\n(Leu starvation)')
ax.legend(loc=2, bbox_to_anchor=(1.2, 1))
fig.savefig('../figures/reviewer_response_fig2.svg')

Content source: rasilab/ferrin_elife_2017

Similar notebooks:

notebook.community | gallery | about