Analyze Differential Features

Find differential TSSs or TSRs from previously generated edgeR or DESeq2 model.

differential_expression(
  experiment,
  data_type = c("tss", "tsr", "tss_features", "tsr_features"),
  comparison_name,
  comparison_type,
  comparison,
  shrink_lfc = FALSE
)

Arguments

experiment	TSRexploreR object.
data_type	Whether the input was generated from TSSs ('tss') or TSRs ('tsr').
comparison_name	The name given to the comparison when stored back into the TSRexploreR object.
comparison_type	For DEseq2, either 'contrast' or 'name'. For edgeR, either 'contrast' or 'coef'.
comparison	For DESeq2, the contrast or name. For edgeR, the coefficients or contrasts.
shrink_lfc	For DESeq2, whether the log2 fold changes are shrunk (TRUE) or not (FALSE).

Value

TSRexploreR object with results for given contrast.

Details

Calculatse the differential TSSs or TSRs for the desired contrast. 'comparison_type' corresponds to the way the comparison will be performed, with edgeR having the 'contrast' and 'coef' options, and DESeq2 having the 'contrast' and 'name' options. The actual contrast is specified with 'comparison', the format of which should match with the option provided to 'comparison_type'. If DESeq2 is used and 'shrink_lfc' is TRUE, apeglm is used to shrink the Log2 fold changes to mitigate the effect size of genes with lower levels of expression. The results for the contrast will be stored back into the TSRexploreR object with the name provided to 'comparison_name'.

See also

fit_de_model to fit DEseq2 or edgeR model to data.

Examples

data(TSSs)
sample_sheet <- data.frame(
  sample_name=c(
    sprintf("S288C_D_%s", seq_len(3)),
    sprintf("S288C_WT_%s", seq_len(3))
  ),
  file_1=rep(NA, 6), file_2=rep(NA, 6),
  condition=c(
    rep("Diamide", 3),
    rep("Untreated", 3)
  )
)

exp <- TSSs %>%
  tsr_explorer(sample_sheet=sample_sheet) %>%
  format_counts(data_type="tss")

# Differential TSSs with DESeq2.
exp <- fit_de_model(exp, data_type="tss", formula=~condition, method="edgeR")

exp <- differential_expression(
  exp, data_type="tss", 
  comparison_name="Diamide_vs_Untreated",
  comparison_type="coef",
  comparison=2
)