Merge branch 'dev' into 'master'

release 0.6.7

See merge request tron/bnt_neoants/splice2neo!122

.Rbuildignore

@@ -9,3 +9,5 @@
 ^pkgdown$
 ^\.github$
 ^codecov\.yml$
+^doc$
+^Meta$

.gitignore

@@ -6,3 +6,5 @@ docs
 fix_issue_*.R
 vignettes/*.html
 vignettes/*.R
+/doc/
+/Meta/

.gitlab-ci.yml

@@ -1,25 +1,29 @@
-image: bioconductor/bioconductor_docker:RELEASE_3_17
+image: bioconductor/bioconductor_docker:RELEASE_3_19
 
 stages:
   - check
   - deploy
 
 testing:
+  artifacts:
+    when: on_failure
+    paths:
+      - testing.log
   stage: check
   script:
-    - R -e 'install.packages("devtools")'
-    - R -e 'devtools::install()'
+    - R -e 'install.packages("devtools", verbose = FALSE, quiet = TRUE)'
+    - R -e 'devtools::install(quiet=TRUE)'
     - R -e 'devtools::check()'
 
 pages:
   stage: deploy
   dependencies:
     - testing
   script:
-    - R -e 'install.packages("devtools")'
-    - R -e 'devtools::install()'
-    - R -e 'install.packages("pkgdown")'
-    - R -e "pkgdown::build_site()"
+    - R -e 'install.packages("devtools", verbose = FALSE, quiet = TRUE)'
+    - R -e 'devtools::install(quiet=TRUE)'
+    - R -e 'install.packages("pkgdown", verbose = FALSE, quiet = TRUE)'
+    - R -e 'pkgdown::build_site()'
     - mkdir -p public
     - cp -r docs/* public
   artifacts:

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: splice2neo
 Title: Aberrant splice junction analysis with associated mutations
-Version: 0.6.6
+Version: 0.6.7
 Authors@R: 
     c(
     person(given = "Jonas",
@@ -11,26 +11,36 @@ Authors@R:
     person(given = "Franziska",
            family = "Lang",
            role = c("aut"),
-           email = "[email protected]")
+           email = "[email protected]"),
+    person(given = "Michelle",
+           family = "Cervante",
+           role = c("aut")),
+    person(given = "Johannes",
+           family = "Hausmann",
+           role = c("aut")),
+    person(given = "Patrick",
+           family = "Sorn",
+           role = c("aut"))
     )
-Description: This package provides functions for the analysis of alternative or 
- aberrant splicing junctions and their creation from or association with 
- somatic mutations. It integrates the output of several tools which predict 
- splicing effects from mutation or RNA-seq data into a common splice junction 
- format based on genomic coordinates. Splice junctions can be annotated with 
- affected transcript sequences, CDS, and resulting peptide sequences.  
+Description: This package provides functions for analyzing alternative splicing 
+ and aberrant splice junctions and their creation from or association with 
+ somatic mutations. It integrates the output of several tools that predict 
+ splicing effects from mutation or RNA-seq data into a unified splice junction 
+ format. Splice junctions can be annotated with affected transcript sequences, 
+ CDS, and resulting peptide sequences.    
 License: MIT + file LICENSE
 Encoding: UTF-8
 LazyData: true
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.2.3
+RoxygenNote: 7.3.1
 Suggests: 
     spelling,
     knitr,
     testthat (>= 3.0.0),
     covr,
     ggbio,
-    rmarkdown
+    rmarkdown,
+    BSgenome,
 Config/testthat/edition: 3
 Language: en-US
 Imports: 
@@ -41,7 +51,7 @@ Imports:
     magrittr,
     tibble,
     stringr (>= 1.5.0),
-    dplyr,
+    dplyr (>= 1.1.1),
     readr,
     tidyr,
     vcfR,
@@ -52,8 +62,7 @@ Imports:
     XVector,
     rlang,
     rtracklayer,
-    BSgenome.Hsapiens.UCSC.hg19,
-    BSgenome
+    BSgenome.Hsapiens.UCSC.hg19
 URL: https://github.com/TRON-Bioinformatics/splice2neo
 Depends: 
     R (>= 2.10)

NAMESPACE

@@ -13,8 +13,11 @@ export(canonical_junctions)
 export(choose_tx)
 export(combine_mut_junc)
 export(exon_in_intron)
+export(filter_irfinder_txt)
+export(format_cispliceai_thresh)
 export(format_pangolin)
 export(format_spliceai)
+export(format_spliceai_thresh)
 export(generate_combined_dataset)
 export(generate_junction_id)
 export(get_exon_inclusion_junction)
@@ -23,27 +26,42 @@ export(get_intronretention_alt_pos)
 export(get_intronretention_genomic_alt_pos)
 export(get_junc_pos)
 export(import_leafcutter_counts)
+export(import_regtools_junc)
 export(import_spladder)
+export(import_stringtie_gtf)
 export(is_canonical)
 export(is_in_rnaseq)
 export(junc2breakpoint)
 export(junc_to_gr)
 export(leafcutter_transform)
 export(map_requant)
 export(modify_tx)
+export(parse_cispliceai_thresh)
 export(parse_gtf)
+export(parse_irfinder_txt)
 export(parse_mmsplice)
 export(parse_pangolin)
 export(parse_spliceai)
+export(parse_spliceai_thresh)
+export(parse_star_sj)
 export(read_requant)
+export(read_suppa_ioe)
+export(regtools_transform)
 export(seq_truncate_nonstop)
 export(spladder_transform)
 export(spladder_transform_format)
+export(stringtie_transform)
+export(stringtie_transform_format)
+export(suppa_import)
+export(suppa_transform)
+export(suppa_transform_format)
 export(transform_for_requant)
 export(transform_leafcutter_counts)
+export(transform_regtools_junc)
 export(unique_junc_mmsplice)
 export(unique_mut_junc)
 import(dplyr)
+import(purrr)
 import(readr)
 import(stringr)
 import(tidyr)

NEWS.md

@@ -1,3 +1,35 @@
+# splice2neo 0.6.7
+
+## Added:
+
+ - Functionality from splice2neo_neoants:
+    - CI-SpliceAI parsing and formatting functions with example data and tests
+    - IRFinder parsing and filtering functions with example data and tests
+    - StringTie parsing and formatting functions
+    - SUPPA2 parsing and formatting functions
+    - STAR parsing and formatting functions with example data and tests
+    - RegTools parsing and formatting functions with example data and tests
+ - add test for empty CI-SpliceAI input file
+ - add more global variables to avoid notes in R CMD check
+
+## Changed:
+
+ - Update CI/CD pipeline for less verbosity during R pkg installations
+ - Return single-row data.frame for empty input in `annotate_mut_effect()`
+ - include 'number_of_supporting_reads` when parsing SplAdder output
+ - generalize `generate_combined_dataset()` to multiple inputs
+ - Update README with list of supported tools
+ - add option consider_intron_retention to `annotate_mut_effect()`
+ - adjust syntax for dplyr >= 1.1.1
+ - remove rarely used pkgs from Imports
+ - simplify code in parse_spliceai_thresh()
+ - minor spell check fixes and doc updates
+
+## Fixed: 
+
+ - fix warnings due to NA's in CI-SpliceAI parsing function
+ - fix use of `generate_combined_dataset()` in vignette
+
 # splice2neo 0.6.6
 
 * add vignette with example workflow
@@ -68,7 +100,7 @@
 
 # splice2neo 0.5.0
 
-* intron rentetion events are now supported by `add_context_seq()`. the resulting context sequence covers the complete intron instead of the exon/intron boundary only. Instead of a the junction position in the cts_seq, the positions are given in form of an interval in the `cts_junc_pos`  column for intron retentions. (0,start_IR, end_IR, end_cts) 
+* intron retention events are now supported by `add_context_seq()`. the resulting context sequence covers the complete intron instead of the exon/intron boundary only. Instead of a the junction position in the cts_seq, the positions are given in form of an interval in the `cts_junc_pos`  column for intron retentions. (0,start_IR, end_IR, end_cts) 
 * `add_peptide()` was adjusted for intron retentions
 * more tests were added for several functions
 * fix small bug in `annotate_spliceai_junction()` that led to annotation with same transcript 
@@ -83,7 +115,7 @@
 
 # splice2neo 0.4.0
 
-* Leafcutterr: The strand information while leafcutter parsing is now retrieved from the cluster id in the count table.
+* Leafcutter: The strand information while leafcutter parsing is now retrieved from the cluster id in the count table.
 * Spladder: previous code changes missed to consider the event type in generation of the junc id for alternative splice sites. This is fixed now.
 * More tests were added
 * An old bug in spladder_transform_mutex_exon was fixed

R/add_peptide.R

@@ -106,23 +106,23 @@ add_peptide <- function(df, cds, flanking_size = 14, bsg = NULL, keep_ranges = F
   # extract context sequence from full peptide and cut before stop codon (*)
   df_positions <- df_sub %>%
     dplyr::mutate(
-    intron_retention = intron_retention,
-    strand = str_sub(df_sub$junc_id, -1),
-    protein = protein %>% as.character(),
-    protein_wt = protein_wt %>% as.character(),
-    frame_shift = frame_shift,
-    junc_in_cds = junc_in_cds,
-    cds_mod_id = stringr::str_c(tx_id, "|", junc_id),
-    cds_length_difference = cds_length_difference,
-    junc_pos_cds = junc_pos_cds,
-    junc_pos_cds_wt = junc_pos_cds_wt,
-    # Get context peptides around junction
-    protein_junc_pos = ceiling(junc_pos_cds / 3),
-    # end for IRs
-    protein_length_difference = ifelse(!frame_shift &
-                                         !intron_retention, cds_length_difference / 3, NA),
-    protein_len = as.numeric(BiocGenerics::width(protein))
-  )
+      intron_retention = intron_retention,
+      strand = str_sub(df_sub$junc_id, -1),
+      protein = protein %>% as.character(),
+      protein_wt = protein_wt %>% as.character(),
+      frame_shift = frame_shift,
+      junc_in_cds = junc_in_cds,
+      cds_mod_id = stringr::str_c(tx_id, "|", junc_id),
+      cds_length_difference = cds_length_difference,
+      junc_pos_cds = junc_pos_cds,
+      junc_pos_cds_wt = junc_pos_cds_wt,
+      # Get context peptides around junction
+      protein_junc_pos = ceiling(junc_pos_cds / 3),
+      # end for IRs
+      protein_length_difference = ifelse(!frame_shift &
+                                           !intron_retention, cds_length_difference / 3, NA),
+      protein_len = as.numeric(BiocGenerics::width(protein))
+    )
 
   df_positions <- df_positions %>%
     is_first_reading_frame() %>%

R/annotate_mut_effect.R

@@ -25,6 +25,9 @@
 #' the relevant genomic position are considered.
 #' If `gene_mapping` is TRUE, potentially affected transcripts from the gene
 #' provided in `effect_df` that cover the relevant genomic positions are considered.
+#' @param consider_intron_retention Indicator whether intron retention events
+#' should be considered as splicing event types when building the resulting
+#' splice junctions.
 #'
 #' @return A data.frame with with additional rows and columns including the
 #' splice junction in the column `junc_id`.
@@ -38,7 +41,18 @@
 #'
 #'
 #' @export
-annotate_mut_effect <- function(effect_df, transcripts, transcripts_gr, gene_mapping = FALSE){
+annotate_mut_effect <- function(effect_df,
+                                transcripts,
+                                transcripts_gr,
+                                gene_mapping = FALSE,
+                                consider_intron_retention = TRUE){
+
+  # choose rules based on intron retention consideration
+  if(consider_intron_retention){
+    effect_rule_table = effect_to_junction_rules
+  } else{
+    effect_rule_table = effect_to_junction_rules_wo_ir
+  }
 
   effect_df <- effect_df  %>%
     mutate(
@@ -55,7 +69,8 @@ annotate_mut_effect <- function(effect_df, transcripts, transcripts_gr, gene_map
     names(var_gr) <- effect_df$effect_index
 
     message("INFO: calculate coordinates of upstream and downstream exons...")
-    # get all possible junctions of by start and end coordinates of upsteam and downstream exons
+
+    # get all possible junctions by start and end coordinates of upstream and downstream exons
     next_junc_df <- next_junctions(var_gr, transcripts, transcripts_gr)
 
     message("INFO: calculate junction coordinates from predicted effect...")
@@ -69,16 +84,34 @@ annotate_mut_effect <- function(effect_df, transcripts, transcripts_gr, gene_map
       filter(
         effect != "DL" | at_end,
         effect != "AL" | at_start
-      ) %>%
+      )
 
-      # add rules
+    # return empty tibble if non of the junctions fulfill the above filters (might be a problem in low mutation burden cases)
+    if(nrow(junc_df) == 0) {
+      junc_df <- junc_df %>%
+        tibble::add_column(
+          "class"= NA,
+          "rule_left"= NA,
+          "rule_right"= NA,
+          "strand_offset"= NA,
+          "coord_1"= NA ,
+          "coord_2"= NA,
+          "left"= NA,
+          "right" = NA,
+          "junc_id"= NA,
+          "tx_junc_id"= NA)
+      return(junc_df)
+    }
+
+    # add rules
+    junc_df <- junc_df %>%
       mutate(
         effect = as.character(effect),
-        # pos = as.integer(POS) + pos_rel
       ) %>%
       left_join(
-        effect_to_junction_rules,
-        by = c("effect")
+        effect_rule_table,
+        by = c("effect"),
+        relationship = "many-to-many"
       ) %>%
 
       # apply rules
@@ -250,6 +283,7 @@ next_junctions <- function(var_gr, transcripts, transcripts_gr){
 }
 
 #' Rules on how a splicing affecting variant creates a junction
+#' @keywords internal
 effect_to_junction_rules <- tribble(
   ~effect, ~event_type,             ~rule_left,        ~rule_right,
   "DL",    "intron retention", "pos",             "pos + strand_offset",
@@ -262,3 +296,16 @@ effect_to_junction_rules <- tribble(
 
   "AG",    "alternative 3prim",  "upstream_end",     "pos",
 )
+
+
+#' Rules on how a splicing affecting variant creates a junction without intron
+#' retention
+#' @keywords internal
+effect_to_junction_rules_wo_ir <- tribble(
+  ~effect, ~event_type,             ~rule_left,        ~rule_right,
+  "DL",    "exon skipping",    "upstream_end",    "downstream_start",
+  "DG",    "alternative 5prim", "pos",             "downstream_start",
+  "AL",    "exon skipping",     "upstream_end",    "downstream_start",
+  "AG",    "alternative 3prim",  "upstream_end",     "pos",
+)
+

R/choose_tx.R

@@ -16,7 +16,7 @@
 #' This function selects transcripts that are more likely to be affected to reduce the amount of junction and transcript combinations.
 #' The function excludes transcripts for which both junction positions are located in an intron. Junctions in a given transcript must either represent an
 #' exon skipping, intron retention, exitron, or alternative splice site event or have both junction positions in an exon. Other junction-transcript combinations are also excluded.
-#' This function may loose relevant or keep irrelevant junction-transcripts in particular in regions with mutliple isoforms with distinct splicing pattern.
+#' This function may loose relevant or keep irrelevant junction-transcripts in particular in regions with multiple isoforms with distinct splicing pattern.
 #'
 #' @examples
 #' junc_df <- tibble::tibble(

R/combine_mut_junc.R

@@ -73,7 +73,9 @@ combine_mut_junc <- function(junc_data_list){
 
   for (df in junc_data_list_names){
     junc_df <- junc_df %>%
-      left_join(df, by = c("mut_id", "tx_id", "junc_id"))
+      left_join(df,
+                by = c("mut_id", "tx_id", "junc_id"),
+                relationship = "many-to-many")
   }
 
   return(junc_df)