In vertesy/Seurat.utils: Seurat.utils - utility functions for Seurat

Vignette for 29 functions in Seurat.Utils.Metadata.R

Updated: 2024/03/26 11:27

Seurat.utils Is a collection of utility functions for Seurat single cell analysis. Functions allow 3D plotting, visualisation of statistics & QC, the automation / multiplexing of plotting, interaction with the Seurat object, etc. Some functionalities require functions from CodeAndRoll and MarkdownReports libraries.> For details, please use the help() function, or browse the source code.

1. Get Metadata Column Names Matching Pattern: getMetaColnames()

Retrieves column names from an object's metadata that match a specified pattern.

 # Assuming `combined.obj` is an object with a meta.data slot
 getMetaColnames()

2. Check if a Column Exists in the Metadata of an S4 Object: metaColnameExists()

This function checks whether a given column exists in the meta.data of a Seurat object.

No examples found.

3. getMetadataColumn: getMetadataColumn()

Retrieves a specified metadata column from a Seurat object and returns it as a named vector.

   # Example usage:
   batch_metadata <- getMetadataColumn(ColName.metadata = "batch", obj = combined.obj, as_numeric = TRUE)

4. Get Unique Levels of a Seurat Object Ident Slot: get_levels_seu()

This function extracts the unique levels present in the 'ident' slot of a Seurat object.

No examples found.

5. Calculate Average Metadata for Seurat Object: calculateAverageMetaData()

Computes specified metrics (e.g., median, mean) for given metadata features across each category

 # Assuming `obj` is a Seurat object with relevant metadata columns:
 results <- calculateAverageMetaData(
   obj = obj,
   meta.features = c("nFeature_RNA", "percent.ribo"),
   ident = "ident_column_name",
   metrics = list("median" = median, "mean" = mean),
   verbose = TRUE
 )
 # This will return a list with data frames containing the median and mean
 # of "nFeature_RNA" and "percent.ribo" for each category in "ident_column_name".

6. getMedianMetric.lsObj: getMedianMetric.lsObj()

Get the median values of different columns in meta.data, can iterate over a list of Seurat objects.

   ls.Seurat <- getMedianMetric.lsObj(
     ls.obj = ls.Seurat, n.datasets = length(ls.Seurat),
     mColname = "percent.mito"
   )

7. getCellIDs.from.meta: getCellIDs.from.meta()

Retrieves cell IDs from a specified metadata column of a Seurat object, where the cell ID matches a provided list of values. The matching operation uses the %in% operator.

   # Example usage:
   getCellIDs.from.meta()

8. Add Metadata to a Seurat object, safely with Checks: addMetaDataSafe()

Wrapper function for AddMetaData that includes additional checks and assertions.

No examples found.

9. seu.add.meta.from.vector: seu.add.meta.from.vector()

Adds a new metadata column to a Seurat object.

   # Example usage:
   combined.obj <- seu.add.meta.from.vector(
     obj = combined.obj,
     metaD.colname = metaD.colname.labeled,
     Label.per.cell = Cl.Label.per.cell
   )

10. Create a Metadata Vector: create.metadata.vector()

This function creates a metadata vector from an input vector and a Seurat object.

 create.metadata.vector(vec = my_vector, obj = my_seurat_object, min.intersect = 50)

11. addMetaFraction: addMetaFraction()

Add a new metadata column to a Seurat object, representing the fraction of a gene set in the transcriptome (expressed as a percentage).

   ls.Seurat[[1]] <- addMetaFraction(col.name = "percent.mito", gene.symbol.pattern = "^MT\\.|^MT-")
   ls.Seurat[[1]] <- addMetaFraction(col.name = "percent.ribo", gene.symbol.pattern = "^RPL|^RPS")
   ls.Seurat[[1]] <- addMetaFraction(col.name = "percent.AC.GenBank", gene.symbol.pattern = "^AC[0-9]{6}\\.")
   ls.Seurat[[1]] <- addMetaFraction(col.name = "percent.AL.EMBL", gene.symbol.pattern = "^AL[0-9]{6}\\.")
   ls.Seurat[[1]] <- addMetaFraction(col.name = "percent.LINC", gene.symbol.pattern = "^LINC0")
   ls.Seurat[[1]] <- addMetaFraction(col.name = "percent.MALAT1", gene.symbol.pattern = "^MALAT1")
   colnames(ls.Seurat[[1]]@meta.data)
   HGA_MarkerGenes <- c(
     "ENO1", "IGFBP2", "WSB1", "DDIT4", "PGK1", "BNIP3", "FAM162A", "TPI1",
     "VEGFA", "PDK1", "PGAM1", "IER2", "FOS", "BTG1", "EPB41L4A-AS1", "NPAS4", "HK2", "BNIP3L",
     "JUN", "ENO2", "GAPDH", "ANKRD37", "ALDOA", "GADD45G", "TXNIP"
   )
   sobj <- addMetaFraction(col.name = "percent.HGA", gene.set = HGA_MarkerGenes, obj = sobj)

12. add.meta.tags: add.meta.tags()

Add metadata tags to a Seurat object dataset.

   ls.Seurat[[1]] <- add.meta.tags(list.of.tags = tags, obj = ls.Seurat[[1]], n = 1)

13. seu.add.meta.from.table: seu.add.meta.from.table()

Add multiple new metadata columns to a Seurat object from a table. #

   combined.obj <- seu.add.meta.from.table()

14. seu.map.and.add.new.ident.to.meta: seu.map.and.add.new.ident.to.meta()

Adds a new metadata column to a Seurat object based on an identity mapping table.

   # Example usage:
   combined.obj <- seu.map.and.add.new.ident.to.meta(
     obj = combined.obj,
     ident.table = clusterIDs.GO.process
   )

15. fix.orig.ident: fix.orig.ident()

Remove the string "filtered_feature_bc_matrix." from "orig.ident". Helper function.

   merged.obj$orig.ident <- fix.orig.ident(obj = merged.obj)
   table(merged.obj$orig.ident)

16. seu.RemoveMetadata: seu.RemoveMetadata()

Remove specified metadata columns from a Seurat object.

 combined.obj <- seu.RemoveMetadata(obj = combined.obj, cols_remove = c("column1", "column2"))

17. Save Metadata from a List of Seurat Objects: saveLsSeuratMetadata()

This function takes a list of Seurat objects, extracts their metadata, and saves it to a file with a specified suffix.

No examples found.

18. Transfer Multiple Metadata Columns Between Two Seurat Objects: transferMetadata()

Transfers specified metadata columns from one Seurat object to another,

 # Assuming `object1` and `object2` are Seurat objects, and you want to transfer
 # metadata columns named 'patientID' and 'treatmentGroup' from `object1` to `object2`:
 object2 <- transferMetadata(
   from = object1, to = object2,
   colname_from = c("patientID", "treatmentGroup")
 )

19. Sample N % of a dataframe (obj@metadata), and return rownames (cell IDs).: sampleNpc()

This function samples a specified percentage of a dataframe (specifically a subset

   # Example usage:
   # Suppose 'MetaData' is a dataframe and 'Pass' is a boolean vector with the same length.
   # The following example will sample 10% of the rows of 'MetaData' where 'Pass' is TRUE.
   sampleNpc(metaDF = MetaData[which(Pass), ], pc = 0.1)

20. Combine Metadata from a list of Seurat objects and Write to TSV: writeCombinedMetadataToTsvFromLsObj()

 # Assuming a list of Seurat objects with meta.data
 mergedMetaData <- writeMetadataToTsv(seuratObjectsList, cols.remove = c("column1", "column2"))

21. Plot Metadata Correlation Heatmap: plotMetadataCorHeatmap()

This function plots a heatmap of metadata correlation values. It accepts a Seurat object

No examples found.

22. Calculate and plot heatmap of cluster medians: heatmap_calc_clust_median()

This function calculates the median of specified variables in a dataframe,

No examples found.

23. plotMetadataMedianFractionBarplot: plotMetadataMedianFractionBarplot()

Generates a barplot of metadata median values.

No examples found.

24. Plot Metadata Category Pie Chart: plotMetadataCategPie()

Generates a pie chart visualizing the distribution of categories within a specified

 plotMetadataCategPie(metacol = "Singlet.status",
                      plot_name = "Singlet Status Distribution",
                      obj = combined.obj,
                      max.categs = 20,
                      both_pc_and_value = TRUE)

25. Rename Azimuth Columns in Seurat Object: renameAzimuthColumns()

Dynamically renames specified metadata columns in a Seurat object, particularly those

 # Assuming `obj` is a Seurat object with metadata columns following the "predicted." pattern:
 obj <- renameAzimuthColumns(obj, ref = "humancortexref", prefix = "azi")
 # This will rename columns like "predicted.class" to "azi.humancortex.class"
 # and include "mapping.score" as "azi.humancortex.mapping.score"

26. Rename Small Categories in Seurat Object Metadata: renameSmallCategories()

This function renames categories within a specified identity column of a

 # Assuming obj is a Seurat object with identity columns "ident1" and "ident2":
 idents <- c("ident1", "ident2")
 obj <- renameSmallCategories(obj, idents = idents)

27. Transfer labels from a reference Seurat object to a query Seurat object: transferLabelsSeurat()

Function to transfer labels from a reference Seurat object to a query Seurat object

 # combined.objX <- transferLabelsSeurat(named_ident = 'RNA_snn_res.0.3.ordered.ManualNames',
 #                                     reference_obj = reference_obj,
 #                                     query_obj = combined.obj)

28. Extract meta.data Column Names Matching a Pattern: .metaColnames()

This function matches the best identity from ident_to_rename to reference_ident in an object,

 # Assuming 'df' is a dataframe with column names "azi.one", "azi.two", "other"
 extract_matching_columns(df, "^azi\\.")

29. Match and Translate Best Identity: matchBestIdentity()

Used for mapping identity columns across objects. This function replaces each

 updated_obj <- matchBestIdentity(my_obj, "origin_identity", "target_identity")

30. Find Best Match: Replace Categories by the Most Frequent Match: .replace_by_most_frequent_categories()

NA

 .replace_by_most_frequent_categories(df = my_data)
 (MXX <- as.tibble(structure(
   c(
     "Adjut", "Adjut", "Yearn", "Adjut", "Dwarf", "Adjut",
     "Dwarf", "Adjut", "Dwarf", "Yearn", "Dwarf", "Dwarf", "Dwarf",
     "Yearn", "Dwarf", "Dwarf", "Dwarf", "Zebra", "Yucca", "Plyer",
     "Blaze", "Blaze", "Dazed", "Blaze", "Swept", "Bold", "Vixen",
     "Bold", "Swept", "Dazed", "Mirth", "Witch", "Vixen", "Dazed",
     "Swept", "Mirth", "Swept", "Vexed", "Query", "Yolk"
   ),
   .Dim = c(20L, 2L), .Dimnames =
     list(NULL, c("RNA_snn_res.0.1.ordered", "RNA_snn_res.0.3.ordered"))
 )))

 z <- .replace_by_most_frequent_categories(df = MXX)
 head(cbind(MXX[, 1], z[, 1]))

vertesy/Seurat.utils documentation built on Dec. 4, 2024, 5:20 p.m.