Computational protocol: Size differences of Arctic marine protists between two climate periods—using the paleoecological record to assess the importance of within‐species trait variation

Similar protocols

Protocol publication

[…] All statistical analyses were performed using R version 3.2.2 (R Core Team, ). For data preparation, we used the packages “plyr” version 1.8.3 (Wickham, ) and “reshape2” version 1.4.1 (Wickham, ). Weighted statistics were performed using the packages “SDMTools” version 1.1–211 (VanDerWal, Falconi, Januchowski, Shoo, & Storlie, ) and “weights” version 0.85 (Pasek & Tahk, ). Plotting was performed using the packages “ggplot2” version 1.0.1 (Wickham, ) and “gridExtra” version 2.0.0 (Auguie, ).For each species, the mean, standard deviation, and standard error of cyst size were calculated and the difference in the mean between the warm and cold periods was tested using a Welch two‐sample t‐test for unequal variances (Welch, ). Normality was assessed visually by plotting histograms of the data distribution of each species (Figure ).Changes in community mean size can be caused by both changes in community structure, that is, species shifts (interspecific changes), and changes in the cyst size of individual species, that is, phenotypic plasticity (intraspecific changes). In order to assess the relative importance of intraspecific vs. interspecific changes, we first calculated the weighted community mean size including the influence of both intra‐ and interspecific changes (x intra+inter) for each period following Equation : (1)x¯intra+inter=∑(∅obs×wobs)∑(wobs)where Øobs are individual observations of cyst diameter and w obs is the weight of each observation given by Equation . (2)wobs=wspeNspewhere N spe is the number of observations of each species and w spe is the weight of each species given by Equation . (3)wspe=NspeNwhere N spe is the number of cysts of each species and N is the total number of cysts (as in Figure b).We then calculated the weighted community mean size excluding intraspecific changes (x inter) for each period following Equation . (4)x¯inter=∑(∅avg×wspe)∑(wspe)where Øavg is the mean cyst size for each species across both periods and w spe is the weight of each species in each period. The relative contribution of intra‐ and interspecific changes to total change can then be calculated following Equations , , . (5)Δx¯intra+inter=x¯(cold)intra+inter−x¯(warm)intra+inter (6)Δx¯inter=x¯(cold)inter−x¯(warm)inter (7)Δx¯intra=Δx¯intra+inter−Δx¯interwhere Δx intra+inter is the size change attributed to the combined effects of intra‐ and interspecific changes, and Δx inter and Δx intra are the contribution of interspecific and intraspecific changes to Δx intra+inter.In addition to the weighted mean, we also calculated the weighted standard deviation and standard error of the mean. A Welch two‐sample weighted t‐test (Bland & Kerry, ) was used to assess the difference in the weighted mean between the two periods. […]

Pipeline specifications

Software tools SDMTools, Ggplot2
Application Miscellaneous