Drop uninformative genes

Source: R/drop_uninformative_genes.r

drop_uninformative_genes drops uninformative genes in order to reduce compute time and noise in subsequent steps. It achieves this through several steps, each of which are optional:

  • Drop non-1:1 orthologs:
    Removes genes that don't have 1:1 orthologs with the output_species ("human" by default).

  • Drop non-varying genes:
    Removes genes that don't vary across cells based on variance deciles.

  • Drop non-differentially expressed genes (DEGs):
    Removes genes that are not significantly differentially expressed across cell-types (multiple DEG methods available).

drop_uninformative_genes(
  exp,
  level2annot,
  mtc_method = "BH",
  adj_pval_thresh = 1e-05,
  convert_orths = FALSE,
  input_species = NULL,
  output_species = "human",
  non121_strategy = "drop_both_species",
  method = "homologene",
  as_sparse = TRUE,
  as_DelayedArray = FALSE,
  return_sce = FALSE,
  no_cores = 1,
  verbose = TRUE,
  ...
)

Arguments

exp

Expression matrix with gene names as rownames.

level2annot

Array of cell types, with each sequentially corresponding a column in the expression matrix.

mtc_method

Multiple-testing correction method used by DGE step. See p.adjust for more details.

adj_pval_thresh

Minimum differential expression significance that a gene must demonstrate across level2annot (i.e. cell types).

convert_orths

If input_species!=output_species and convert_orths=TRUE, will drop genes without 1:1 output_species orthologs and then convert exp gene names to those of output_species.

input_species

Which species the gene names in exp come from. See list_species for all available species.

output_species

Which species' genes names to convert exp to. See list_species for all available species.

non121_strategy

How to handle genes that don't have 1:1 mappings between input_species:output_species. Options include:

  • "drop_both_species" or "dbs" or 1 :
    Drop genes that have duplicate mappings in either the input_species or output_species
    (DEFAULT).

  • "drop_input_species" or "dis" or 2 :
    Only drop genes that have duplicate mappings in the input_species.

  • "drop_output_species" or "dos" or 3 :
    Only drop genes that have duplicate mappings in the output_species.

  • "keep_both_species" or "kbs" or 4 :
    Keep all genes regardless of whether they have duplicate mappings in either species.

  • "keep_popular" or "kp" or 5 :
    Return only the most "popular" interspecies ortholog mappings. This procedure tends to yield a greater number of returned genes but at the cost of many of them not being true biological 1:1 orthologs.

  • "sum","mean","median","min" or "max" :
    When gene_df is a matrix and gene_output="rownames", these options will aggregate many-to-one gene mappings (input_species-to-output_species) after dropping any duplicate genes in the output_species.

method

R package to use for gene mapping:

  • "gprofiler" : Slower but more species and genes.

  • "homologene" : Faster but fewer species and genes.

  • "babelgene" : Faster but fewer species and genes. Also gives consensus scores for each gene mapping based on a several different data sources.

as_sparse

Convert exp to sparse matrix.

as_DelayedArray

Convert exp to DelayedArray for scalable processing.

return_sce

Whether to return the filtered results as an expression matrix or a SingleCellExperiment.

no_cores

Number of cores to parallelise across. Set to NULL to automatically optimise.

verbose

Print messages. #' @inheritParams orthogene::convert_orthologs

...

Arguments passed on to orthogene::convert_orthologs

gene_df

Data object containing the genes (see gene_input for options on how the genes can be stored within the object).
Can be one of the following formats:

  • matrix :
    A sparse or dense matrix.

  • data.frame :
    A data.frame, data.table. or tibble.

  • codelist :
    A list or character vector.

Genes, transcripts, proteins, SNPs, or genomic ranges can be provided in any format (HGNC, Ensembl, RefSeq, UniProt, etc.) and will be automatically converted to gene symbols unless specified otherwise with the ... arguments.
Note: If you set method="homologene", you must either supply genes in gene symbol format (e.g. "Sox2") OR set standardise_genes=TRUE.

gene_input

Which aspect of gene_df to get gene names from:

  • "rownames" :
    From row names of data.frame/matrix.

  • "colnames" :
    From column names of data.frame/matrix.

  • <column name> :
    From a column in gene_df, e.g. "gene_names".

gene_output

How to return genes. Options include:

  • "rownames" :
    As row names of gene_df.

  • "colnames" :
    As column names of gene_df.

  • "columns" :
    As new columns "input_gene", "ortholog_gene" (and "input_gene_standard" if standardise_genes=TRUE) in gene_df.

  • "dict" :
    As a dictionary (named list) where the names are input_gene and the values are ortholog_gene.

  • "dict_rev" :
    As a reversed dictionary (named list) where the names are ortholog_gene and the values are input_gene.

standardise_genes

If TRUE AND gene_output="columns", a new column "input_gene_standard" will be added to gene_df containing standardised HGNC symbols identified by gorth.

drop_nonorths

Drop genes that don't have an ortholog in the output_species.

agg_fun

Aggregation function passed to aggregate_mapped_genes. Set to NULL to skip aggregation step (default).

mthreshold

Maximum number of ortholog names per gene to show. Passed to gorth. Only used when method="gprofiler" (DEFAULT : Inf).

sort_rows

Sort gene_df rows alphanumerically.

gene_map

A data.frame that maps the current gene names to new gene names. This function's behaviour will adapt to different situations as follows:

  • gene_map=<data.frame> :
    When a data.frame containing the gene key:value columns (specified by input_col and output_col, respectively) is provided, this will be used to perform aggregation/expansion.

  • gene_map=NULL and input_species!=output_species :
    A gene_map is automatically generated by map_orthologs to perform inter-species gene aggregation/expansion.

  • gene_map=NULL and input_species==output_species :
    A gene_map is automatically generated by map_genes to perform within-species gene gene symbol standardization and aggregation/expansion.

input_col

Column name within gene_map with gene names matching the row names of X.

output_col

Column name within gene_map with gene names that you wish you map the row names of X onto.

Value

exp Expression matrix with gene names as row names.

Examples

cortex_mrna <- ewceData::cortex_mrna()
#> see ?ewceData and browseVignettes('ewceData') for documentation
#> loading from cache
# Use only a subset of genes to keep the example quick
cortex_mrna$exp <- cortex_mrna$exp[1:300, ]

## Convert orthologs at the same time
exp2_orth <- drop_uninformative_genes(
    exp = cortex_mrna$exp,
    level2annot = cortex_mrna$annot$level2class,
    input_species = "mouse"
)
#> Check 300Check 3005
#> Warning: sctSpecies_origin not provided. Setting to 'mouse' by default.
#> 1 core(s) assigned as workers (3 reserved).
#> Converting to sparse matrix.
#> Checking for non-expressed genes.
#> Checking for cells with no expressed genes.
#> DGE:: Limma...
#> 3 / 300 genes dropped @ DGE adj_pval_thresh < 1e-05
#> Time difference of 0.07441974 secs