Example: Multiple experiments
Leiv Rønneberg
05/09/2022
Example_screen.RmdThe synergyscreen provides a work flow for data from big
drug combination screens, where multiple drugs are tested in combination
on multiple cell lines. It takes as input a list of experiments, each
entry being a list containing the necessary elements needed for a call
to the main regression function bayesynergy.
Melanoma cell line example
Included in the package is the result of a synergyscreen
run of 583 drug combinations on the A-375 human melanoma cell line from
ONeil et al. (2016). The
synergyscreen object is a list with two entries, a
dataframe with parameter estimates from each experiment, and a list
entitled failed – containing experiments that either failed
completely to process, or had an unsatisfactory fit.
library(bayesynergy)
data("ONeil_A375")
length(ONeil_A375$failed)## [1] 2We see that the dataset has two experiments that failed to process,
during an initial run of synergyscreen. There’s a multitude
of reasons why an experiment might fail to process, it could be an input
error, initialization problems or problems with the parallel
processing.
The entries of failed are themselves lists, each
containing the necessary information to process through the
bayesynergy function
failed_experiment = ONeil_A375$failed[[1]]
names(failed_experiment)## [1] "y" "x" "drug_names" "experiment_ID"## viability L778123 MK-4827
## [1,] 0.759 0.325 0.223
## [2,] 0.755 0.325 0.775
## [3,] 0.548 0.325 2.750
## [4,] 0.307 0.325 10.000
## [5,] 0.787 0.800 0.223
## [6,] 0.820 0.800 0.775We can rerun experiments that failed to process, by simply passing
the returned synergyscreen object back into the function.
Note that we turn of the default options of saving each fit and plotting
everything, and set method = "vb" indicating we use
variational inference to fit the model.
fit_screen = synergyscreen(ONeil_A375, save_raw = F, save_plots = F, parallel = F,
bayesynergy_params = list(method = "vb"))## Chain 1: ------------------------------------------------------------
## Chain 1: EXPERIMENTAL ALGORITHM:
## Chain 1: This procedure has not been thoroughly tested and may be unstable
## Chain 1: or buggy. The interface is subject to change.
## Chain 1: ------------------------------------------------------------
## Chain 1:
## Chain 1:
## Chain 1:
## Chain 1: Gradient evaluation took 0.000402 seconds
## Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 4.02 seconds.
## Chain 1: Adjust your expectations accordingly!
## Chain 1:
## Chain 1:
## Chain 1: Begin eta adaptation.
## Chain 1: Iteration: 1 / 250 [ 0%] (Adaptation)
## Chain 1: Iteration: 50 / 250 [ 20%] (Adaptation)
## Chain 1: Iteration: 100 / 250 [ 40%] (Adaptation)
## Chain 1: Iteration: 150 / 250 [ 60%] (Adaptation)
## Chain 1: Iteration: 200 / 250 [ 80%] (Adaptation)
## Chain 1: Iteration: 250 / 250 [100%] (Adaptation)
## Chain 1: ------------------------------------------------------------
## Chain 1: EXPERIMENTAL ALGORITHM:
## Chain 1: This procedure has not been thoroughly tested and may be unstable
## Chain 1: or buggy. The interface is subject to change.
## Chain 1: ------------------------------------------------------------
## Chain 1:
## Chain 1:
## Chain 1:
## Chain 1: Gradient evaluation took 0.000306 seconds
## Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 3.06 seconds.
## Chain 1: Adjust your expectations accordingly!
## Chain 1:
## Chain 1:
## Chain 1: Begin eta adaptation.
## Chain 1: Iteration: 1 / 250 [ 0%] (Adaptation)
## Chain 1: Iteration: 50 / 250 [ 20%] (Adaptation)
## Chain 1: Iteration: 100 / 250 [ 40%] (Adaptation)
## Chain 1: Iteration: 150 / 250 [ 60%] (Adaptation)
## Chain 1: Iteration: 200 / 250 [ 80%] (Adaptation)
## Chain 1: Iteration: 250 / 250 [100%] (Adaptation)
## Chain 1: ------------------------------------------------------------
## Chain 1: EXPERIMENTAL ALGORITHM:
## Chain 1: This procedure has not been thoroughly tested and may be unstable
## Chain 1: or buggy. The interface is subject to change.
## Chain 1: ------------------------------------------------------------
## Chain 1:
## Chain 1:
## Chain 1:
## Chain 1: Gradient evaluation took 0.000265 seconds
## Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 2.65 seconds.
## Chain 1: Adjust your expectations accordingly!
## Chain 1:
## Chain 1:
## Chain 1: Begin eta adaptation.
## Chain 1: Iteration: 1 / 250 [ 0%] (Adaptation)
## Chain 1: Iteration: 50 / 250 [ 20%] (Adaptation)
## Chain 1: Iteration: 100 / 250 [ 40%] (Adaptation)
## Chain 1: Iteration: 150 / 250 [ 60%] (Adaptation)
## Chain 1: Iteration: 200 / 250 [ 80%] (Adaptation)
## Chain 1: Iteration: 250 / 250 [100%] (Adaptation)
## Chain 1: Success! Found best value [eta = 0.1].
## Chain 1:
## Chain 1: Begin stochastic gradient ascent.
## Chain 1: iter ELBO delta_ELBO_mean delta_ELBO_med notes
## Chain 1: 100 -4860.788 1.000 1.000
## Chain 1: 200 -10126.267 0.760 1.000
## Chain 1: 300 -5561.405 0.780 0.821
## Chain 1: 400 -5530.223 0.587 0.821
## Chain 1: 500 -4317.703 0.525 0.520
## Chain 1: 600 -7811.948 0.512 0.520
## Chain 1: 700 -4960.627 0.521 0.520
## Chain 1: 800 -4268.776 0.476 0.520
## Chain 1: 900 -4957.297 0.439 0.447
## Chain 1: 1000 -3895.517 0.422 0.447
## Chain 1: 1100 -3266.061 0.342 0.281
## Chain 1: 1200 -2129.613 0.343 0.281
## Chain 1: 1300 -2725.341 0.283 0.273
## Chain 1: 1400 -1851.887 0.329 0.281
## Chain 1: 1500 -2584.757 0.330 0.284
## Chain 1: 1600 -2055.410 0.311 0.273
## Chain 1: 1700 -2347.692 0.266 0.258
## Chain 1: 1800 -2168.106 0.258 0.258
## Chain 1: 1900 -1911.069 0.257 0.258
## Chain 1: 2000 -1876.216 0.232 0.219
## Chain 1: 2100 -1597.373 0.230 0.219
## Chain 1: 2200 -2008.067 0.197 0.205
## Chain 1: 2300 -1361.956 0.223 0.205
## Chain 1: 2400 -1502.685 0.185 0.175
## Chain 1: 2500 -1292.627 0.173 0.163
## Chain 1: 2600 -2192.369 0.188 0.163
## Chain 1: 2700 -1110.858 0.273 0.175
## Chain 1: 2800 -1500.580 0.291 0.205
## Chain 1: 2900 -1066.382 0.318 0.260
## Chain 1: 3000 -1151.814 0.323 0.260
## Chain 1: 3100 -1195.934 0.310 0.260
## Chain 1: 3200 -1049.637 0.303 0.260
## Chain 1: 3300 -1004.046 0.260 0.163
## Chain 1: 3400 -1514.632 0.285 0.260
## Chain 1: 3500 -892.361 0.338 0.337
## Chain 1: 3600 -1069.318 0.314 0.260
## Chain 1: 3700 -799.804 0.250 0.260
## Chain 1: 3800 -968.228 0.241 0.174
## Chain 1: 3900 -1135.559 0.215 0.165
## Chain 1: 4000 -959.012 0.226 0.174
## Chain 1: 4100 -829.659 0.238 0.174
## Chain 1: 4200 -750.800 0.235 0.174
## Chain 1: 4300 -832.245 0.240 0.174
## Chain 1: 4400 -868.118 0.211 0.165
## Chain 1: 4500 -681.331 0.168 0.165
## Chain 1: 4600 -654.109 0.156 0.156
## Chain 1: 4700 -593.368 0.132 0.147
## Chain 1: 4800 -708.854 0.131 0.147
## Chain 1: 4900 -657.695 0.124 0.105
## Chain 1: 5000 -590.246 0.117 0.105
## Chain 1: 5100 -604.318 0.104 0.102
## Chain 1: 5200 -665.249 0.103 0.098
## Chain 1: 5300 -745.381 0.104 0.102
## Chain 1: 5400 -543.667 0.137 0.108
## Chain 1: 5500 -590.189 0.117 0.102
## Chain 1: 5600 -605.695 0.116 0.102
## Chain 1: 5700 -507.219 0.125 0.108
## Chain 1: 5800 -612.470 0.126 0.108
## Chain 1: 5900 -557.804 0.128 0.108
## Chain 1: 6000 -553.142 0.117 0.098
## Chain 1: 6100 -681.902 0.134 0.108
## Chain 1: 6200 -573.949 0.143 0.172
## Chain 1: 6300 -540.277 0.139 0.172
## Chain 1: 6400 -443.817 0.123 0.172
## Chain 1: 6500 -578.591 0.139 0.188
## Chain 1: 6600 -413.764 0.176 0.189
## Chain 1: 6700 -431.288 0.161 0.188
## Chain 1: 6800 -494.692 0.156 0.188
## Chain 1: 6900 -463.815 0.153 0.188
## Chain 1: 7000 -432.226 0.160 0.188
## Chain 1: 7100 -398.188 0.149 0.128
## Chain 1: 7200 -440.302 0.140 0.096
## Chain 1: 7300 -421.724 0.138 0.096
## Chain 1: 7400 -447.638 0.122 0.085
## Chain 1: 7500 -400.582 0.111 0.085
## Chain 1: 7600 -415.116 0.074 0.073
## Chain 1: 7700 -431.546 0.074 0.073
## Chain 1: 7800 -440.883 0.063 0.067
## Chain 1: 7900 -437.082 0.058 0.058
## Chain 1: 8000 -427.175 0.053 0.044
## Chain 1: 8100 -451.642 0.050 0.044
## Chain 1: 8200 -344.162 0.071 0.044
## Chain 1: 8300 -336.808 0.069 0.038
## Chain 1: 8400 -338.868 0.064 0.035
## Chain 1: 8500 -360.947 0.058 0.035
## Chain 1: 8600 -378.831 0.059 0.038
## Chain 1: 8700 -302.918 0.081 0.047
## Chain 1: 8800 -349.337 0.092 0.054
## Chain 1: 8900 -348.412 0.091 0.054
## Chain 1: 9000 -293.931 0.107 0.061
## Chain 1: 9100 -344.485 0.117 0.133
## Chain 1: 9200 -308.051 0.097 0.118
## Chain 1: 9300 -320.628 0.099 0.118
## Chain 1: 9400 -303.323 0.104 0.118
## Chain 1: 9500 -334.784 0.107 0.118
## Chain 1: 9600 -293.070 0.117 0.133
## Chain 1: 9700 -311.054 0.098 0.118
## Chain 1: 9800 -280.948 0.095 0.107
## Chain 1: 9900 -280.061 0.095 0.107
## Chain 1: 10000 -288.767 0.080 0.094
## Chain 1: Informational Message: The maximum number of iterations is reached! The algorithm may not have converged.
## Chain 1: This variational approximation is not guaranteed to be meaningful.
## Chain 1:
## Chain 1: Drawing a sample of size 1000 from the approximate posterior...
## Chain 1: COMPLETED.
## Chain 1: ------------------------------------------------------------
## Chain 1: EXPERIMENTAL ALGORITHM:
## Chain 1: This procedure has not been thoroughly tested and may be unstable
## Chain 1: or buggy. The interface is subject to change.
## Chain 1: ------------------------------------------------------------
## Chain 1:
## Chain 1:
## Chain 1:
## Chain 1: Gradient evaluation took 0.000316 seconds
## Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 3.16 seconds.
## Chain 1: Adjust your expectations accordingly!
## Chain 1:
## Chain 1:
## Chain 1: Begin eta adaptation.
## Chain 1: Iteration: 1 / 250 [ 0%] (Adaptation)
## Chain 1: Iteration: 50 / 250 [ 20%] (Adaptation)
## Chain 1: Iteration: 100 / 250 [ 40%] (Adaptation)
## Chain 1: Iteration: 150 / 250 [ 60%] (Adaptation)
## Chain 1: Iteration: 200 / 250 [ 80%] (Adaptation)
## Chain 1: Iteration: 250 / 250 [100%] (Adaptation)
## Chain 1: ------------------------------------------------------------
## Chain 1: EXPERIMENTAL ALGORITHM:
## Chain 1: This procedure has not been thoroughly tested and may be unstable
## Chain 1: or buggy. The interface is subject to change.
## Chain 1: ------------------------------------------------------------
## Chain 1:
## Chain 1:
## Chain 1:
## Chain 1: Gradient evaluation took 0.000259 seconds
## Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 2.59 seconds.
## Chain 1: Adjust your expectations accordingly!
## Chain 1:
## Chain 1:
## Chain 1: Begin eta adaptation.
## Chain 1: Iteration: 1 / 250 [ 0%] (Adaptation)
## Chain 1: Iteration: 50 / 250 [ 20%] (Adaptation)
## Chain 1: Iteration: 100 / 250 [ 40%] (Adaptation)
## Chain 1: Iteration: 150 / 250 [ 60%] (Adaptation)
## Chain 1: Iteration: 200 / 250 [ 80%] (Adaptation)
## Chain 1: Iteration: 250 / 250 [100%] (Adaptation)
## Chain 1: Success! Found best value [eta = 0.1].
## Chain 1:
## Chain 1: Begin stochastic gradient ascent.
## Chain 1: iter ELBO delta_ELBO_mean delta_ELBO_med notes
## Chain 1: 100 -1740.278 1.000 1.000
## Chain 1: 200 -883.230 0.985 1.000
## Chain 1: 300 -1174.491 0.739 0.970
## Chain 1: 400 -852.065 0.649 0.970
## Chain 1: 500 -959.049 0.542 0.378
## Chain 1: 600 -821.924 0.479 0.378
## Chain 1: 700 -1108.896 0.448 0.259
## Chain 1: 800 -788.226 0.443 0.378
## Chain 1: 900 -752.925 0.399 0.259
## Chain 1: 1000 -760.918 0.360 0.259
## Chain 1: 1100 -736.928 0.263 0.248
## Chain 1: 1200 -673.625 0.175 0.167
## Chain 1: 1300 -669.215 0.151 0.112
## Chain 1: 1400 -794.585 0.129 0.112
## Chain 1: 1500 -624.534 0.145 0.158
## Chain 1: 1600 -598.237 0.133 0.094
## Chain 1: 1700 -679.851 0.119 0.094
## Chain 1: 1800 -575.220 0.097 0.094
## Chain 1: 1900 -613.729 0.098 0.094
## Chain 1: 2000 -616.162 0.098 0.094
## Chain 1: 2100 -651.031 0.100 0.094
## Chain 1: 2200 -567.211 0.105 0.120
## Chain 1: 2300 -744.950 0.128 0.148
## Chain 1: 2400 -560.266 0.145 0.148
## Chain 1: 2500 -606.584 0.126 0.120
## Chain 1: 2600 -563.640 0.129 0.120
## Chain 1: 2700 -532.267 0.123 0.076
## Chain 1: 2800 -627.207 0.120 0.076
## Chain 1: 2900 -477.556 0.145 0.148
## Chain 1: 3000 -536.006 0.155 0.148
## Chain 1: 3100 -542.696 0.151 0.148
## Chain 1: 3200 -527.549 0.139 0.109
## Chain 1: 3300 -507.293 0.120 0.076
## Chain 1: 3400 -502.618 0.088 0.076
## Chain 1: 3500 -495.865 0.081 0.059
## Chain 1: 3600 -505.255 0.076 0.040
## Chain 1: 3700 -515.305 0.072 0.029
## Chain 1: 3800 -492.210 0.061 0.029
## Chain 1: 3900 -508.379 0.033 0.029
## Chain 1: 4000 -494.316 0.025 0.028
## Chain 1: 4100 -464.703 0.030 0.029
## Chain 1: 4200 -481.386 0.031 0.032
## Chain 1: 4300 -573.361 0.043 0.032
## Chain 1: 4400 -487.944 0.059 0.035
## Chain 1: 4500 -453.269 0.066 0.047
## Chain 1: 4600 -455.245 0.064 0.047
## Chain 1: 4700 -450.021 0.063 0.047
## Chain 1: 4800 -434.506 0.062 0.036
## Chain 1: 4900 -461.114 0.065 0.058
## Chain 1: 5000 -460.708 0.062 0.058
## Chain 1: 5100 -436.157 0.061 0.056
## Chain 1: 5200 -400.749 0.067 0.058
## Chain 1: 5300 -481.626 0.067 0.058
## Chain 1: 5400 -439.459 0.060 0.058
## Chain 1: 5500 -458.603 0.056 0.056
## Chain 1: 5600 -435.841 0.061 0.056
## Chain 1: 5700 -426.247 0.062 0.056
## Chain 1: 5800 -442.995 0.062 0.056
## Chain 1: 5900 -405.141 0.066 0.056
## Chain 1: 6000 -425.115 0.070 0.056
## Chain 1: 6100 -473.757 0.075 0.088
## Chain 1: 6200 -395.644 0.086 0.093
## Chain 1: 6300 -409.774 0.073 0.052
## Chain 1: 6400 -402.019 0.065 0.047
## Chain 1: 6500 -385.313 0.065 0.047
## Chain 1: 6600 -418.965 0.068 0.047
## Chain 1: 6700 -394.287 0.072 0.063
## Chain 1: 6800 -385.328 0.070 0.063
## Chain 1: 6900 -418.699 0.069 0.063
## Chain 1: 7000 -361.886 0.080 0.080
## Chain 1: 7100 -358.030 0.071 0.063
## Chain 1: 7200 -368.963 0.054 0.043
## Chain 1: 7300 -354.685 0.055 0.043
## Chain 1: 7400 -394.876 0.063 0.063
## Chain 1: 7500 -405.902 0.061 0.063
## Chain 1: 7600 -370.953 0.063 0.063
## Chain 1: 7700 -392.190 0.062 0.054
## Chain 1: 7800 -423.907 0.067 0.075
## Chain 1: 7900 -348.140 0.081 0.075
## Chain 1: 8000 -351.915 0.066 0.054
## Chain 1: 8100 -410.714 0.079 0.075
## Chain 1: 8200 -372.081 0.087 0.094
## Chain 1: 8300 -391.642 0.088 0.094
## Chain 1: 8400 -345.930 0.091 0.094
## Chain 1: 8500 -373.147 0.095 0.094
## Chain 1: 8600 -371.905 0.086 0.075
## Chain 1: 8700 -341.658 0.090 0.089
## Chain 1: 8800 -381.708 0.093 0.104
## Chain 1: 8900 -347.128 0.081 0.100
## Chain 1: 9000 -364.200 0.085 0.100
## Chain 1: 9100 -328.459 0.081 0.100
## Chain 1: 9200 -318.881 0.074 0.089
## Chain 1: 9300 -351.735 0.078 0.093
## Chain 1: 9400 -328.591 0.072 0.089
## Chain 1: 9500 -352.075 0.071 0.089
## Chain 1: 9600 -362.543 0.074 0.089
## Chain 1: 9700 -335.201 0.073 0.082
## Chain 1: 9800 -331.359 0.064 0.070
## Chain 1: 9900 -345.581 0.058 0.067
## Chain 1: 10000 -355.387 0.056 0.067
## Chain 1: Informational Message: The maximum number of iterations is reached! The algorithm may not have converged.
## Chain 1: This variational approximation is not guaranteed to be meaningful.
## Chain 1:
## Chain 1: Drawing a sample of size 1000 from the approximate posterior...
## Chain 1: COMPLETED.
## Chain 1: ------------------------------------------------------------
## Chain 1: EXPERIMENTAL ALGORITHM:
## Chain 1: This procedure has not been thoroughly tested and may be unstable
## Chain 1: or buggy. The interface is subject to change.
## Chain 1: ------------------------------------------------------------
## Chain 1:
## Chain 1:
## Chain 1:
## Chain 1: Gradient evaluation took 0.000321 seconds
## Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 3.21 seconds.
## Chain 1: Adjust your expectations accordingly!
## Chain 1:
## Chain 1:
## Chain 1: Begin eta adaptation.
## Chain 1: Iteration: 1 / 250 [ 0%] (Adaptation)
## Chain 1: Iteration: 50 / 250 [ 20%] (Adaptation)
## Chain 1: Iteration: 100 / 250 [ 40%] (Adaptation)
## Chain 1: Iteration: 150 / 250 [ 60%] (Adaptation)
## Chain 1: Iteration: 200 / 250 [ 80%] (Adaptation)
## Chain 1: Iteration: 250 / 250 [100%] (Adaptation)
## Chain 1: Success! Found best value [eta = 0.1].
## Chain 1:
## Chain 1: Begin stochastic gradient ascent.
## Chain 1: iter ELBO delta_ELBO_mean delta_ELBO_med notes
## Chain 1: 100 -840.096 1.000 1.000
## Chain 1: 200 -480.433 0.874 1.000
## Chain 1: 300 -639.272 0.666 0.749
## Chain 1: 400 -821.495 0.555 0.749
## Chain 1: 500 -578.721 0.528 0.420
## Chain 1: 600 -895.460 0.499 0.420
## Chain 1: 700 -1271.501 0.470 0.354
## Chain 1: 800 -445.018 0.643 0.420
## Chain 1: 900 -517.479 0.587 0.354
## Chain 1: 1000 -410.798 0.554 0.354
## Chain 1: 1100 -557.552 0.481 0.296
## Chain 1: 1200 -477.736 0.423 0.263
## Chain 1: 1300 -430.359 0.409 0.263
## Chain 1: 1400 -405.340 0.393 0.263
## Chain 1: 1500 -605.161 0.384 0.263
## Chain 1: 1600 -323.551 0.436 0.263
## Chain 1: 1700 -421.450 0.429 0.260
## Chain 1: 1800 -382.934 0.254 0.232
## Chain 1: 1900 -398.451 0.243 0.232
## Chain 1: 2000 -353.770 0.230 0.167
## Chain 1: 2100 -391.602 0.213 0.126
## Chain 1: 2200 -370.174 0.202 0.110
## Chain 1: 2300 -423.666 0.204 0.126
## Chain 1: 2400 -413.249 0.200 0.126
## Chain 1: 2500 -347.253 0.186 0.126
## Chain 1: 2600 -361.776 0.103 0.101
## Chain 1: 2700 -354.107 0.082 0.097
## Chain 1: 2800 -322.251 0.082 0.097
## Chain 1: 2900 -339.203 0.083 0.097
## Chain 1: 3000 -373.944 0.080 0.093
## Chain 1: 3100 -256.972 0.116 0.093
## Chain 1: 3200 -273.861 0.116 0.093
## Chain 1: 3300 -356.905 0.127 0.093
## Chain 1: 3400 -289.865 0.147 0.099
## Chain 1: 3500 -282.025 0.131 0.093
## Chain 1: 3600 -274.295 0.130 0.093
## Chain 1: 3700 -312.355 0.140 0.099
## Chain 1: 3800 -304.688 0.133 0.093
## Chain 1: 3900 -282.474 0.136 0.093
## Chain 1: 4000 -326.621 0.140 0.122
## Chain 1: 4100 -227.638 0.138 0.122
## Chain 1: 4200 -293.713 0.154 0.135
## Chain 1: 4300 -238.639 0.154 0.135
## Chain 1: 4400 -261.939 0.140 0.122
## Chain 1: 4500 -282.993 0.144 0.122
## Chain 1: 4600 -215.268 0.173 0.135
## Chain 1: 4700 -182.044 0.179 0.183
## Chain 1: 4800 -225.611 0.196 0.193
## Chain 1: 4900 -216.872 0.192 0.193
## Chain 1: 5000 -230.731 0.184 0.193
## Chain 1: 5100 -205.898 0.153 0.183
## Chain 1: 5200 -266.060 0.153 0.183
## Chain 1: 5300 -241.290 0.140 0.121
## Chain 1: 5400 -198.088 0.153 0.183
## Chain 1: 5500 -211.846 0.152 0.183
## Chain 1: 5600 -198.791 0.127 0.121
## Chain 1: 5700 -192.726 0.112 0.103
## Chain 1: 5800 -205.411 0.099 0.066
## Chain 1: 5900 -201.744 0.097 0.066
## Chain 1: 6000 -200.403 0.092 0.066
## Chain 1: 6100 -199.061 0.080 0.065
## Chain 1: 6200 -191.360 0.062 0.062
## Chain 1: 6300 -181.480 0.057 0.054
## Chain 1: 6400 -169.791 0.042 0.054
## Chain 1: 6500 -196.522 0.049 0.054
## Chain 1: 6600 -172.586 0.056 0.054
## Chain 1: 6700 -182.417 0.059 0.054
## Chain 1: 6800 -208.015 0.065 0.054
## Chain 1: 6900 -152.542 0.099 0.069
## Chain 1: 7000 -196.517 0.121 0.123
## Chain 1: 7100 -146.268 0.155 0.136
## Chain 1: 7200 -193.864 0.175 0.139
## Chain 1: 7300 -185.046 0.174 0.139
## Chain 1: 7400 -147.455 0.193 0.224
## Chain 1: 7500 -140.992 0.184 0.224
## Chain 1: 7600 -167.028 0.186 0.224
## Chain 1: 7700 -152.199 0.190 0.224
## Chain 1: 7800 -127.861 0.197 0.224
## Chain 1: 7900 -127.626 0.161 0.190
## Chain 1: 8000 -151.039 0.154 0.156
## Chain 1: 8100 -167.047 0.129 0.155
## Chain 1: 8200 -103.028 0.167 0.155
## Chain 1: 8300 -162.585 0.198 0.156
## Chain 1: 8400 -143.814 0.186 0.155
## Chain 1: 8500 -114.402 0.207 0.156
## Chain 1: 8600 -148.756 0.215 0.190
## Chain 1: 8700 -122.816 0.226 0.211
## Chain 1: 8800 -149.826 0.225 0.211
## Chain 1: 8900 -145.014 0.228 0.211
## Chain 1: 9000 -135.265 0.220 0.211
## Chain 1: 9100 -131.859 0.213 0.211
## Chain 1: 9200 -104.993 0.176 0.211
## Chain 1: 9300 -158.963 0.174 0.211
## Chain 1: 9400 -105.710 0.211 0.231
## Chain 1: 9500 -108.546 0.188 0.211
## Chain 1: 9600 -121.626 0.176 0.180
## Chain 1: 9700 -101.924 0.174 0.180
## Chain 1: 9800 -82.120 0.180 0.193
## Chain 1: 9900 -95.033 0.190 0.193
## Chain 1: 10000 -99.660 0.188 0.193
## Chain 1: Informational Message: The maximum number of iterations is reached! The algorithm may not have converged.
## Chain 1: This variational approximation is not guaranteed to be meaningful.
## Chain 1:
## Chain 1: Drawing a sample of size 1000 from the approximate posterior...
## Chain 1: COMPLETED.
References
ONeil, Jennifer, Yair Benita, Igor Feldman, Melissa Chenard, Brian
Roberts, Yaping Liu, Jing Li, et al. 2016. “An Unbiased Oncology
Compound Screen to Identify Novel Combination Strategies.”
Molecular Cancer Therapeutics 15 (6): 1155–62. https://doi.org/10.1158/1535-7163.mct-15-0843.