forked from AlexsLemonade/OpenPBTA-analysis
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Update fusion-summary to include union of biospecimen IDs in fusion c…
…allers (AlexsLemonade#478) * Add note about putative oncogenic fusions single caller * Add in biospecimens in either caller Change to notebook for documentation purposes * Run notebook instead Also +x * Remove Rscript * Minor typo, formatting fixes * Add in the original caller files to modules at a glance * Add in 'missing fusions' To better match former behavior * Will embryonal step pass in CI? * Revert "Will embryonal step pass in CI?" This reverts commit b81379d. * Skip ependymoma steps in CI * Forgot the variable in CI * Forgot to replace NA with 0 * Apply @jashapiro right_join suggestion * Apply suggestions from code review Co-Authored-By: jashapiro <jashapiro@gmail.com> Co-authored-by: jashapiro <jashapiro@gmail.com>
- Loading branch information
1 parent
e6a165e
commit 0e642ef
Showing
10 changed files
with
4,280 additions
and
1,640 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file was deleted.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,186 @@ | ||
| --- | ||
| title: "Generate Fusion Summary Files" | ||
| output: html_notebook | ||
| author: Daniel Miller (D3b) and Jaclyn Taroni (CCDL) | ||
| date: January 2020 | ||
| params: | ||
| is_ci: 0 | ||
| --- | ||
|
|
||
| Generate fusion files specifically for consumption by molecular subtyping analyses | ||
|
|
||
| ## Set up | ||
|
|
||
| ```{r} | ||
| # if running in CI, we need to skip the EPN steps | ||
| if (params$is_ci == 0) running_in_ci <- FALSE | ||
| if (params$is_ci == 1) running_in_ci <- TRUE | ||
| ``` | ||
|
|
||
| ### Libraries and functions | ||
|
|
||
| ```{r} | ||
| library(tidyverse) | ||
| ``` | ||
|
|
||
| ```{r} | ||
| #' Generate filtered fusion frame | ||
| #' @param df Unfiltered fusion data frame | ||
| #' @param bioid List of biospecimen IDs | ||
| #' @param fuses List of explicit fusion names | ||
| #' @param genes List of gene names | ||
| #' @return the filtered fusion data frame | ||
| filterFusion <- function(df, bioid, fuses, genes) { | ||
| if (!missing(bioid)) { | ||
| df <- filter(df, Sample %in% bioid) | ||
| } | ||
| if (!missing(fuses) & !missing(genes)) { | ||
| df <- filter(df, FusionName %in% fuses | | ||
| Gene1A %in% genes | | ||
| Gene2A %in% genes | | ||
| Gene1B %in% genes | | ||
| Gene2B %in% genes) | ||
| } else if (!missing(fuses)) { | ||
| df <- filter(df, FusionName %in% fuses) | ||
| } else if (!missing(genes)) { | ||
| df <- filter(df, | ||
| Gene1A %in% genes | | ||
| Gene2A %in% genes | | ||
| Gene1B %in% genes | | ||
| Gene2B %in% genes) | ||
| } | ||
| return(df %>% select(Sample, FusionName)) | ||
| } | ||
| #' Generate matrix with fusion counts | ||
| #' @param fuseDF Filtered fusion data frame | ||
| #' @param bioid List of biospecimen IDs that should be included in final table | ||
| #' @return Data frame that contains fusion counts | ||
| prepareOutput <- function(fuseDF, bioid) { | ||
| fuseDF %>% | ||
| reshape2::dcast(Sample ~ FusionName) %>% | ||
| right_join(data.frame(Sample = bioid)) %>% | ||
| replace(is.na(.), 0) %>% | ||
| rename(Kids_First_Biospecimen_ID = Sample) | ||
| } | ||
| ``` | ||
|
|
||
| ### Read in data | ||
|
|
||
| ```{r} | ||
| dataDir <- file.path("..", "..", "data") | ||
| #' The putative oncogenic fusion file is what we'll use to check for the | ||
| #' presence or absence of the fusions. | ||
| putativeOncogenicDF <- | ||
| read_tsv(file.path(dataDir, "pbta-fusion-putative-oncogenic.tsv")) | ||
| #' However, some biospecimens are not represented in this filtered, prioritized | ||
| #' file but *are* present in the original files -- this will cause them to be | ||
| #' "missing" in the final files for consumption which could mislead analysts. | ||
| arribaDF <- read_tsv(file.path(dataDir, "pbta-fusion-arriba.tsv.gz")) | ||
| starfusionDF <- read_tsv(file.path(dataDir, "pbta-fusion-starfusion.tsv.gz")) | ||
| ``` | ||
|
|
||
| ### Output | ||
|
|
||
| ```{r} | ||
| resultsDir <- "results" | ||
| if (!dir.exists(resultsDir)) { | ||
| dir.create(resultsDir) | ||
| } | ||
| ependFile <- file.path(resultsDir, "fusion_summary_ependymoma_foi.tsv") | ||
| embryFile <- file.path(resultsDir, "fusion_summary_embryonal_foi.tsv") | ||
| ``` | ||
|
|
||
| ## Fusions and genes of interest | ||
|
|
||
| Taken from [`AlexsLemonade/OpenPBTA-analysis#245`](https://github.com/AlexsLemonade/OpenPBTA-analysis/issues/245) and [`AlexsLemonade/OpenPBTA-analysis#251`](https://github.com/AlexsLemonade/OpenPBTA-analysis/issues/251), respectively. | ||
|
|
||
| ```{r} | ||
| #' **Filters** | ||
| #' | ||
| #' *Fusions Filters* | ||
| #' 1: Exact match a list of fusions common in Ependymoma tumors | ||
| ependFuses <- c( | ||
| "C11orf95--MAML2", | ||
| "C11orf95--RELA", | ||
| "C11orf95--YAP1", | ||
| "LTBP3--RELA", | ||
| "PTEN--TAS2R1", | ||
| "YAP1--FAM118B", | ||
| "YAP1--MAMLD1", | ||
| "YAP1--MAMLD2" | ||
| ) | ||
| ependGenes <- c( | ||
| "RELA" | ||
| ) | ||
| #' 2: Exact match a list of fusions common in Embryonal tumors | ||
| #' as well as fusions containing a particular gene with any other gene | ||
| embryFuses <- c( | ||
| "CIC--NUTM1", | ||
| "MN1--BEND2", | ||
| "MN1--CXXC5" | ||
| ) | ||
| embryGenes <- c( | ||
| "FOXR2", | ||
| "MN1", | ||
| "TTYH1" | ||
| ) | ||
| ``` | ||
|
|
||
| ### Filter putative oncogenic fusions list | ||
|
|
||
| ```{r} | ||
| allFuseEpend <- filterFusion(df = putativeOncogenicDF, | ||
| fuses = ependFuses, | ||
| genes = ependGenes) | ||
| allFuseEmbry <- filterFusion(df = putativeOncogenicDF, | ||
| fuses = embryFuses, | ||
| genes = embryGenes) | ||
| ``` | ||
|
|
||
| Get the biospecimen IDs that are present in *either* caller file (Arriba, STARFusion). | ||
| The fusions in the putative oncogenic fusion file can be retained even if they are not in both callers: https://github.com/AlexsLemonade/OpenPBTA-analysis/blob/8fba1753608d8ac0aa3d5d7d63c480b8f00ff0e9/analyses/fusion_filtering/04-project-specific-filtering.Rmd#L242 | ||
| We use the putative oncogenic file here, therefore any sample that is in either file but does not have a fusion that is relevant to the subtyping tickets is not _missing_ but instead has no evidence of the relevant fusions. | ||
|
|
||
| ```{r} | ||
| specimensUnion<- union(arribaDF$tumor_id, starfusionDF$tumor_id) | ||
| ``` | ||
|
|
||
| #### Write non-MB, non-ATRT embryonal fusions to file | ||
|
|
||
| ```{r} | ||
| allFuseEmbry <- allFuseEmbry %>% | ||
| prepareOutput(specimensUnion) | ||
| ``` | ||
|
|
||
| ```{r} | ||
| # Are there any missing fusions? | ||
| setdiff(embryFuses, colnames(allFuseEmbry)) | ||
| ``` | ||
|
|
||
| ```{r} | ||
| allFuseEmbry %>% | ||
| mutate( | ||
| `CIC--NUTM1` = 0, | ||
| `MN1--BEND2` = 0 | ||
| ) %>% | ||
| write_tsv(embryFile) | ||
| ``` | ||
|
|
||
| #### Write ependymoma fusions to file | ||
|
|
||
| ```{r} | ||
| if (!running_in_ci) { | ||
| allFuseEpend %>% | ||
| prepareOutput(specimensUnion) %>% | ||
| mutate( | ||
| `C11orf95--YAP1` = 0, | ||
| `LTBP3--RELA` = 0, | ||
| `PTEN--TAS2R1` = 0, | ||
| `YAP1--MAMLD2` = 0 | ||
| ) %>% | ||
| write_tsv(ependFile) | ||
| } | ||
| ``` |
2,102 changes: 2,102 additions & 0 deletions
2,102
analyses/fusion-summary/01-fusion-summary.nb.html
Large diffs are not rendered by default.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.