The experimentally observed percentages of the five types of assembled channels expressed following a 1:1 injection of G375R mutant and WT cRNA did not differ significantly from the theoretical percentages calculated assuming random assembly of G375R mutant and WT subunits. Simulation was used to determine if the percentages of observed and predicted channel types (Fig. 5 B and Fig. 1) were significantly different by examining whether the observed numbers of channels of each type in the experimental group of 33 assembled channels (Fig. 5 A) were significantly different from the expected numbers based on random assembly of subunits (this figure), as it is the numbers that determine the percentages. The expected numbers could not be calculated directly because of the random assembly of subunits but could be specified in terms of simulated discrete probability distributions, one for each of the five channel types (see Materials and methods). This figure plots the five discrete probability distributions, where the specific channel type (stoichiometry) for each distribution is indicated by a schematic, where black circles are WT subunits and blue circles are mutant subunits. Each one of the five discrete probability distributions plots the probability of observing 0, 1, 2, 3…33 assembled channels of the type indicated for that distribution in a group of 33 channels randomly assembled from mutant and WT subunits. The five discrete probability distributions were generated by first simulating 106 groups of 33 randomly assembled channels (see Materials and methods and Fig. 1) and then binning the number of each type of assembled channel in each group of 33 channels into five frequency histograms, one for each channel type. The bins in each frequency histogram were then divided by 106, the total counts in each frequency histogram, to normalize each frequency histograms to discrete probability distributions with an area of 1.0. To illustrate with some examples, the topmost discrete probability distribution, which gives the probabilities of observing different numbers of WT channels in a group of 33 assembled channels, indicates that the probability of observing 0 WT channels in a group of 33 randomly assembled channels is 0.119. The probability of observing 1 WT channel is 0.262, and the probability of observing 2 WT channels is 0.278, etc. Thus, 11.9, 26.2, and 27.8% of the groups of 33 assembled channels would have 0, 1, or 2 WT channels, respectively. The arrow at 2.06 WT channels indicates the mean number of WT channels per group of 33 assembled channels from the analysis of the 106 groups. The expected mean number of WT channels per group of 33 channels can also be calculated directly from the percentages in Fig. 1, where 6.25% of the 33 assembled channels on average would be WT, giving a mean of 2.06 WT assembled channels per group of 33 assemble channels (0.0625 × 33 = 2.06). The discrete probability distributions for each of the four other types of assembled channels are also presented and can be interpreted in a similar manner. Our experimental observations of the number of assembled channels of each type in the experimental group of 33 assembled channels (Fig. 4 D; and Fig. 5, A and B) are indicated by the locations of the red histogram bars on each abscissa of the discrete probability distributions. The amplitude of the red bars gives the probability of observing the indicated number of channels of that type assuming a random assembly of subunits. For example, the probability for our observation of 1 WT channel in the experimental group of 33 assembled channels was 0.262, and the probabilities for our observations of 10, 11, 7, and 4 assembled channels for the other indicated types of assembled channels in the experimental group of 33 assembled channels were 0.118, 0.128, 0.147, and 0.0961, respectively. All of these probabilities for the observed numbers of assembled channels of each type were >0.05, and none of the observed numbers of assembled channel types fell within the 0.05 summed cumulative probability of the left and right tails of the discrete probability distributions, indicating that the experimental observations for the number of the five types of assembled channels were not significantly different (Colquhoun, 1971) from the theoretical predictions based on an assumption of random subunit assembly. A lack of significant difference does not exclude that there may be underlying differences between experimental and theoretical predictions, which could be revealed by larger sample sizes.
Figure S4.

The experimentally observed percentages of the five types of assembled channels expressed following a 1:1 injection of G375R mutant and WT cRNA did not differ significantly from the theoretical percentages calculated assuming random assembly of G375R mutant and WT subunits. Simulation was used to determine if the percentages of observed and predicted channel types (Fig. 5 B and Fig. 1) were significantly different by examining whether the observed numbers of channels of each type in the experimental group of 33 assembled channels (Fig. 5 A) were significantly different from the expected numbers based on random assembly of subunits (this figure), as it is the numbers that determine the percentages. The expected numbers could not be calculated directly because of the random assembly of subunits but could be specified in terms of simulated discrete probability distributions, one for each of the five channel types (see Materials and methods). This figure plots the five discrete probability distributions, where the specific channel type (stoichiometry) for each distribution is indicated by a schematic, where black circles are WT subunits and blue circles are mutant subunits. Each one of the five discrete probability distributions plots the probability of observing 0, 1, 2, 3…33 assembled channels of the type indicated for that distribution in a group of 33 channels randomly assembled from mutant and WT subunits. The five discrete probability distributions were generated by first simulating 106 groups of 33 randomly assembled channels (see Materials and methods and Fig. 1) and then binning the number of each type of assembled channel in each group of 33 channels into five frequency histograms, one for each channel type. The bins in each frequency histogram were then divided by 106, the total counts in each frequency histogram, to normalize each frequency histograms to discrete probability distributions with an area of 1.0. To illustrate with some examples, the topmost discrete probability distribution, which gives the probabilities of observing different numbers of WT channels in a group of 33 assembled channels, indicates that the probability of observing 0 WT channels in a group of 33 randomly assembled channels is 0.119. The probability of observing 1 WT channel is 0.262, and the probability of observing 2 WT channels is 0.278, etc. Thus, 11.9, 26.2, and 27.8% of the groups of 33 assembled channels would have 0, 1, or 2 WT channels, respectively. The arrow at 2.06 WT channels indicates the mean number of WT channels per group of 33 assembled channels from the analysis of the 106 groups. The expected mean number of WT channels per group of 33 channels can also be calculated directly from the percentages in Fig. 1, where 6.25% of the 33 assembled channels on average would be WT, giving a mean of 2.06 WT assembled channels per group of 33 assemble channels (0.0625 × 33 = 2.06). The discrete probability distributions for each of the four other types of assembled channels are also presented and can be interpreted in a similar manner. Our experimental observations of the number of assembled channels of each type in the experimental group of 33 assembled channels (Fig. 4 D; and Fig. 5, A and B) are indicated by the locations of the red histogram bars on each abscissa of the discrete probability distributions. The amplitude of the red bars gives the probability of observing the indicated number of channels of that type assuming a random assembly of subunits. For example, the probability for our observation of 1 WT channel in the experimental group of 33 assembled channels was 0.262, and the probabilities for our observations of 10, 11, 7, and 4 assembled channels for the other indicated types of assembled channels in the experimental group of 33 assembled channels were 0.118, 0.128, 0.147, and 0.0961, respectively. All of these probabilities for the observed numbers of assembled channels of each type were >0.05, and none of the observed numbers of assembled channel types fell within the 0.05 summed cumulative probability of the left and right tails of the discrete probability distributions, indicating that the experimental observations for the number of the five types of assembled channels were not significantly different (Colquhoun, 1971) from the theoretical predictions based on an assumption of random subunit assembly. A lack of significant difference does not exclude that there may be underlying differences between experimental and theoretical predictions, which could be revealed by larger sample sizes.

This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal