combineCommonResults: Combine SingleR results with common genes
In SingleR: Reference-Based Single-Cell RNA-Seq Annotation

Description Usage Arguments Details Value Author(s) See Also Examples

Combine results from multiple runs of classifySingleR (usually against different references) into a single DataFrame. This assumes that each run of classifySingleR was performed using a common set of marker genes.

1	combineCommonResults(results)

results

A list of DataFrame prediction results as returned by classifySingleR when run on each reference separately.

Here, the strategy is to performed classification separately within each reference, then collating the results to choose the label with the highest score across references. For each cell, we identify the reference with the highest score across all of its labels. The “combined label” is then defined as the label assigned to that cell in the highest-scoring reference. (The same logic is also applied to the first and pruned labels, if those are available.)

Each result should be generated from training sets that use a common set of genes during classification, i.e., common.genes should be the same in the trained argument to each classifySingleR call. This is because the scores are not comparable across results if they were generated from different sets of genes. It is also for this reason that we use the highest score prior to fine-tuning, even if it does not correspond to the score of the fine-tuned label.

It is highly unlikely that this function will be called directly by the end-user. Users are advised to use the multi-reference mode of SingleR and related functions, which will take care of the use of a common set of genes before calling this function to combine results across references.

A DataFrame is returned containing the annotation statistics for each cell or cluster (row). This mimics the output of classifySingleR and contains the following fields:

scores, a numeric matrix of correlations formed by combining the equivalent matrices from results.
labels, a character vector containing the per-cell combined label across references.
references, an integer vector specifying the reference from which the combined label was derived.
orig.results, a DataFrame containing results.

It may also contain first.labels and pruned.labels if these were also present in results.

The metadata contains common.genes, a character vector of the common genes that were used across all references in results; and label.origin, a DataFrame specifying the reference of origin for each label in scores.

Jared Andrews, Aaron Lun

SingleR and classifySingleR, for generating predictions to use in results.

combineRecomputedResults, for another approach to combining predictions.

# Making up data (using one reference to seed another).
ref <- .mockRefData(nreps=8)
ref1 <- ref[,1:2%%2==0]
ref2 <- ref[,1:2%%2==1]
ref2$label <- tolower(ref2$label)

test <- .mockTestData(ref1)

# Applying classification with SingleR's multi-reference mode.
ref1 <- scuttle::logNormCounts(ref1)
ref2 <- scuttle::logNormCounts(ref2)
test <- scuttle::logNormCounts(test)

pred <- SingleR(test, list(ref1, ref2), labels=list(ref1$label, ref2$label))
pred[,1:5] # Only viewing the first 5 columns for visibility.