Single sample analysis

Descriptive analyses can be applied on each repertoire individually. These analyses include the ones described in the Data exploration section in addition to other metrics.

In all the following functions, the sample_id of interest needs to be specified in the sampleName parameter. If not specified, the first sample_id in the dataset will be analyzed.

Repertoire statistics

Metadata statistics

The plotIndStatistics() function allows the positioning of the sample of interest within the whole dataset, based on a set of statistic metrics. Boxplots are drawn on the whole dataset values, on which the sample of interest is highlighted as a red cross. This allows the comparison of the sample to the whole dataset, and the detection of sample outliers.

To plot statistics in the metaData slot, the stat parameter should be set to statistics in the metaData slot.

plotIndStatistics(x = RepSeqData, 
                  stat = "metadata",  
                  level = "aaClone")
#> Plot the first sample in the dataset: tripod-30-813

Figure 1: Boxplots showing metadata statistics calculated on all the dataset samples. The sample of interest is represented by a red cross.

V and J gene usages

V-J combination usages can be visualized in a heatmap using the plotIndGeneUsage() function. The parameter level allows to choose whether to compute gene usages at the ntClone or aaClone level. V and J gene usages are plotted separately at the top and right side of the heatmap, allowing a better view of the individual gene distribution.

plotIndGeneUsage(x = RepSeqData,
                 level = "aaClone",
                 sampleName = NULL)
#> Plot the first sample in the dataset: tripod-30-813

Figure 5: Heatmap of V and J combination usage in one selected sample. Frequencies, represented by the color scale, are scaled column-wise. Barplots at the top and right side of the heatmap show the usage of each gene across the row or the column for the V and J genes, respectively.

Repertoire diversity

The plotIndStatistics() function can also be used to plot diversity indices by setting the stat parameter to diversity.

plotIndStatistics(x = RepSeqData, 
                  stat = "diversity",  
                  level = "aaClone")
#> Plot the first sample in the dataset: tripod-30-813

Figure 2: Boxplots showing different diversity indices calculated on all the dataset samples. The sample of interest is represented by a red cross.

Clonal distribution

Tree map

To visualize the repertoire structure and the top clone distribution in a repertoire, plotIndMap() can be used to plot a circular treemap in which each circle represents a unique clone, and the circle size corresponds to the clone count. Users can choose the proportion of the top clones to plot using the prop parameter. We recommend to plot up to the top 10,000 clones, as higher values might generate difficult-to-interpret visuals.

plotIndMap(x = RepSeqData, 
           sampleName=  NULL, 
           level = "aaClone", 
           prop = 0.01)
#> Plot the first sample in the dataset: tripod-30-813

Figure 3: A circular treemap was generated for one single sample and shows the structure of the top 1% of aaClones in the sample repertoire. Each circle represents a unique aaClone, and the circle size corresponds to the aaClone count.

Intervals

The plotIndIntervals() function can be used to quantitatively analyze the above plot, by evaluating the clonal distribution per intervals of counts or fractions in a single sample.

plotIndIntervals(x = RepSeqData, 
                 level = "aaClone", 
                 sampleName = "tripod-30-813", 
                 interval_scale = "frequency")

Figure 4: Barplots showing the distribution (upper plot) and the cumulative frequency (lower plot) of aaClone within each frequency interval for one selected sample.