{"type":"doc","content":[{"type":"extension","attrs":{"layout":"default","extensionType":"com.atlassian.confluence.macro.core","extensionKey":"toc","parameters":{"macroParams":{"style":{"value":"none"}},"macroMetadata":{"macroId":{"value":"0ebff2ce-0d1b-43a6-b4e0-c0009ae8b9d6"},"schemaVersion":{"value":"1"},"title":"Table of Contents"}},"localId":"c6c14fef-9691-4603-aa57-76ffad109a55"}},{"type":"paragraph"},{"type":"heading","attrs":{"level":1},"content":[{"text":"Overview","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"While alpha diversity examines differences within treatment groups, beta diversity measures the similarity (or dissimilarity) ","type":"text"},{"text":"between","type":"text","marks":[{"type":"strong"}]},{"text":" samples and groups.","type":"text"}]},{"type":"paragraph","content":[{"text":"Each variable is plotted on ","type":"text"},{"text":"principal coordinates analysis (PCoA)","type":"text","marks":[{"type":"link","attrs":{"href":"https://archetypalecology.wordpress.com/2018/02/19/principal-coordinates-analysis-pcoa-in-r/"}}]},{"text":" plots, to examine the variance between samples based on a dissimilarity matrix. A detailed explanation of PCoA and other ordination methods can be seen here: ","type":"text"},{"type":"inlineCard","attrs":{"url":"http://albertsenlab.org/ampvis2-ordination/"}},{"text":" ","type":"text"}]},{"type":"paragraph","content":[{"text":"Sample distance can be measured using 3 distance-based ordination methods. These methods are:","type":"text"}]},{"type":"orderedList","attrs":{"order":1},"content":[{"type":"listItem","content":[{"type":"paragraph","content":[{"text":"Bray–Curtis dissimilarity","type":"text","marks":[{"type":"strong"},{"type":"link","attrs":{"href":"https://www.statisticshowto.com/bray-curtis-dissimilarity/"}}]},{"text":" measures the fraction of overabundant counts.","type":"text"}]}]}]},{"type":"paragraph","content":[{"text":"Sorenson, T. (1948) “A method of establishing groups of equal amplitude in plant sociology based on similarity of species content.” Kongelige Danske Videnskabernes Selskab 5.1-34: 4-7.","type":"text","marks":[{"type":"code"}]}]},{"type":"orderedList","attrs":{"order":2},"content":[{"type":"listItem","content":[{"type":"paragraph","content":[{"text":"Cao index","type":"text","marks":[{"type":"strong"},{"type":"link","attrs":{"href":"https://www.sciencedirect.com/science/article/abs/pii/S0043135496003223"}}]},{"text":" is a minimally biased index for high beta diversity and variable sampling intensity. Chao index tries to take into account the number of unseen species pairs.","type":"text"}]}]}]},{"type":"paragraph","content":[{"text":"Cao, Y., Bark, A. W., & Williams, W. P. (1997). Analysing benthic macroinvertebrate community changes along a pollution gradient: a framework for the development of biotic indices. Water Research, 31(4), 884-892.","type":"text","marks":[{"type":"code"}]}]},{"type":"orderedList","attrs":{"order":3},"content":[{"type":"listItem","content":[{"type":"paragraph","content":[{"text":"Jaccard similarity index","type":"text","marks":[{"type":"strong"},{"type":"link","attrs":{"href":"https://www.statology.org/jaccard-similarity/"}}]},{"text":" measures the fraction of unique features, regardless of abundance..","type":"text"}]}]}]},{"type":"paragraph","content":[{"text":"Jaccard, P. (1908). “Nouvellesrecherches sur la distribution florale.” Bull. Soc. V and. Sci. Nat., (44):223-270.","type":"text","marks":[{"type":"code"}]}]},{"type":"paragraph"},{"type":"paragraph"},{"type":"heading","attrs":{"level":1},"content":[{"text":"5a. Choosing a variable to analyse","type":"text"}]},{"type":"paragraph","content":[{"text":" ","type":"text"}]},{"type":"paragraph","content":[{"text":"In the Alpha Diversity section, you already imported your nfcore/ampliseq results and converted them to an ampvis2 object. To have another quick look at your ampvis object:","type":"text"}]},{"type":"codeBlock","content":[{"text":"ampvisdata","type":"text"}]},{"type":"paragraph","content":[{"text":" ","type":"text"}]},{"type":"paragraph","content":[{"text":"As in the Alpha Diversity section, you need to choose a variable to work with.","type":"text"}]},{"type":"paragraph","content":[{"text":"You can view your variables as column names in your samples_table:","type":"text"}]},{"type":"codeBlock","content":[{"text":"colnames(samples_table)","type":"text"}]},{"type":"paragraph","content":[{"text":" ","type":"text"}]},{"type":"paragraph","content":[{"text":"Now enter the column name of the variable you want to analyse:","type":"text"}]},{"type":"codeBlock","content":[{"text":"group <- \"Nose_size\"","type":"text"}]},{"type":"paragraph","content":[{"text":" ","type":"text"}]},{"type":"heading","attrs":{"level":2},"content":[{"text":"Ordering your variable","type":"text"}]},{"type":"paragraph","content":[{"text":" ","type":"text"}]},{"type":"paragraph","content":[{"text":"And once again you’ll need to order the groups in your variable. See the ‘Ordering your variable’ section in the Alpha diversity section for more details. ","type":"text"}]},{"type":"paragraph","content":[{"text":"Choose how you want to order your groups:","type":"text"}]},{"type":"codeBlock","content":[{"text":"lev <- c(\"Small\", \"Medium\", \"Big\")","type":"text"}]},{"type":"paragraph","content":[{"text":"Then run the following to apply the levels to your data:","type":"text"}]},{"type":"codeBlock","content":[{"text":"ampvisdata$metadata[[group]] <- factor(ampvisdata$metadata[[group]], levels = lev)","type":"text"}]},{"type":"paragraph","content":[{"text":" ","type":"text"}]},{"type":"heading","attrs":{"level":1},"content":[{"text":"5b. PCoA plots and statistics","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"The overview section described (with links and references) the ordination methods that can be used to estimate and plot beta diversity.","type":"text"}]},{"type":"paragraph","content":[{"text":"Briefly, these are: ","type":"text"},{"text":"Bray–Curtis dissimilarity","type":"text","marks":[{"type":"strong"}]},{"text":", ","type":"text"},{"text":"Cao index","type":"text","marks":[{"type":"strong"}]},{"text":", and ","type":"text"},{"text":"Jaccard similarity index","type":"text","marks":[{"type":"strong"}]},{"text":". Each of these has strengths and weaknesses. It's up to you, the researcher, to explore the literature and decide which is the best index to use for your data.","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"Choose the ordination method you want to use to estimate and plot beta diversity.","type":"text"}]},{"type":"paragraph","content":[{"text":"Bray–Curtis dissimilarity is used by default (","type":"text"},{"text":"\"bray\"","type":"text","marks":[{"type":"code"}]},{"text":") Change this to ","type":"text"},{"text":"\"cao\"","type":"text","marks":[{"type":"code"}]},{"text":" for Cao index, or ","type":"text"},{"text":"\"jaccard\"","type":"text","marks":[{"type":"code"}]},{"text":" for Jaccard similarity index.","type":"text"}]},{"type":"codeBlock","content":[{"text":"index <- \"bray\"","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"Now create the PCoA plot.","type":"text"}]},{"type":"codeBlock","content":[{"text":"p <- amp_ordinate(ampvisdata, type = \"pcoa\", transform = \"none\", distmeasure = index, sample_color_by = group, sample_point_size = 3, sample_colorframe = TRUE) +\nscale_color_manual(values=c(\"Red\", \"Green\", \"Blue\")) +\nscale_fill_manual(values=c(\"Red\", \"Green\", \"Blue\")) +\ntheme(text = element_text(size = 16))\np","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"Save your plot as a 300dpi (i.e. publication quality) tiff or pdf file:","type":"text"}]},{"type":"codeBlock","content":[{"text":"tiff_exp <- paste0(\"PCoA_beta_div_\", group, \"\", index, \"_samples.tif\")\nggsave(file = tiff_exp, dpi = 300, compression = \"lzw\", device = \"tiff\", plot = p, width = 20, height = 20, units = \"cm\")\n\npdf_exp <- paste0(\"PCoA_beta_div_\", group, \"_\", index,, \"_samples.pdf\")\nggsave(file = pdf_exp, device = \"pdf\", plot = p, width = 20, height = 20, units = \"cm\")","type":"text"}]},{"type":"paragraph"},{"type":"heading","attrs":{"level":2},"content":[{"text":"Statistical analysis","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"To compare the overall differences between groups within your chosen variable, a ","type":"text"},{"text":"Permutational Multivariate Analysis of Variance (PERMANOVA)","type":"text","marks":[{"type":"link","attrs":{"href":"https://archetypalecology.wordpress.com/2018/02/21/permutational-multivariate-analysis-of-variance-permanova-in-r-preliminary/"}}]},{"text":" test can be performed and similarly a pairwise PERMANOVA test can be performed to compare differences between each group.","type":"text"}]},{"type":"paragraph","content":[{"text":"The R-squared value represents the percentage of variance explained by the examined groups. E.g. if R = 0.23 then 23% of the total diversity is explained by groupwise differences.","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"PERMANOVA:","type":"text","marks":[{"type":"strong"}]}]},{"type":"codeBlock","content":[{"text":"# Need to remove rows (from ASV abundance table) with all 0 counts first\nasvmatrix <- ampvisdata$abund\nasvmatrix <- asvmatrix[rowSums(asvmatrix) > 0, ]\n# Also need to transpose (samples need to be as rows, asv's as columns)\nasvmatrix <- t(asvmatrix)\n# Then generate pairwise distance matrix\nsampdist <- vegdist(asvmatrix, method=\"bray\")\n# Subset samples table by samples in ampvis2 object (in case some were filtered out)\nsamples_table_filt <- samples_table[samples_table$ID %in% ampvisdata$metadata$ID, ]\n# Use adonis function (vegan package: \"Permutational Multivariate Analysis of Variance Using Distance Matrices\") to run PERMANOVA on distances\npathotype.adonis <- adonis2(sampdist ~ get(group), data = samples_table_filt)\n# Output the r squared and p values as variables\nr2 <- pathotype.adonis$R2[1]\npval <- pathotype.adonis$`Pr(>F)`[1]","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"PERMANOVA R squared =","type":"text"}]},{"type":"codeBlock","content":[{"text":"round(r2, 2)","type":"text"}]},{"type":"paragraph","content":[{"text":"PERMANOVA significance (p) =","type":"text"}]},{"type":"codeBlock","content":[{"text":"pval","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"Pairwise PERMANOVA:","type":"text","marks":[{"type":"strong"}]}]},{"type":"paragraph","content":[{"text":"Calculate the pairwise PERMANOVA. This is a bit complex, as each group within the variable has to be compared to each other group in a variety of ways. Code comments (#) explain what each line of code does.","type":"text"}]},{"type":"codeBlock","content":[{"text":"# The combn function creates every combination of provided elements\n# Below it takes all group names and combines them pairwise (2)\n# Creates a matrix where each column = a combination\ncomb_pair <- data.frame(combn(unique(samples_table_filt[[group]]),2))\n# Convert scores to relative abundance\n# Use sweep function to divide (\"/\") each column (2) by its total (colSums)\ncomm <- sweep(ampvisdata$abund,2,colSums(ampvisdata$abund),\"/\")\n# Using adonis function (vegan package: \"Permutational Multivariate Analysis of Variance Using Distance Matrices\")\ntabstat_adonis <- c()\n# Loop through each pair (i.e. column in 'comb_pair')\nfor (i in 1:ncol(comb_pair)) {\n # Pull out pair data\n # From samples table\n samples_table_SB_pair <- samples_table_filt[samples_table_filt[[group]] %in% comb_pair[[i]], ]\n # From ASV matrix\n asvmatrix_pair <- comm[samples_table_SB_pair$ID]\n # Transpose\n asvmatrix_pair <- asvmatrix_pair[rowSums(asvmatrix_pair) > 0, ]\n # Also need to transpose (samples need to be as rows, asv's as columns)\n asvmatrix_pair <- t(asvmatrix_pair)\n # Then generate pairwise distance matrix\n sampdist_pair <- vegdist(asvmatrix_pair, method=\"bray\")\n # Use anosim (vegan): Analysis of Similarities\n x2 <- adonis2(sampdist_pair ~ get(group), data = samples_table_SB_pair)\n # Pull out just r squared and p value\n r2_adonis <- x2$R2[1]\n pval_adonis <- x2$`Pr(>F)`[1]\n # Combine into data frame\n tabstat_adonis <- cbind(tabstat_adonis, c(r2_adonis, pval_adonis))\n # Name vector with group combinations\n colnames(tabstat_adonis)[i] <- paste(comb_pair[[i]], collapse=' Vs ')\n}\nrow.names(tabstat_adonis) <- c(\"R squared\", \"p\")","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"Now output these results as a table.","type":"text"}]},{"type":"codeBlock","content":[{"text":"tabstat_adonis <- t(tabstat_adonis)\ntabstat_adonis <- data.frame(tabstat_adonis)\ntabstat_adonis","type":"text"}]},{"type":"paragraph","content":[{"text":"You can export this table as a csv file:","type":"text"}]},{"type":"codeBlock","content":[{"text":"write_csv(tabstat_adonis, paste0(\"beta_div_pairwise_PERMANOVA\", group, \"_\", index, \"_samples_.csv\"))","type":"text"}]},{"type":"paragraph"},{"type":"paragraph","content":[{"text":"Go to the next section: Community structure","type":"text"}]},{"type":"paragraph"},{"type":"paragraph"},{"type":"paragraph"}],"version":1}

Browser not supported