From 3380a05a1883e6c5ec509ce764af975b1620fc02 Mon Sep 17 00:00:00 2001
From: Wolf <22243650+Wolfffff@users.noreply.github.com>
Date: Tue, 21 Nov 2023 09:34:05 -0600
Subject: [PATCH] fix: Fixing workflow for single-end sequencing (#70)

* Correcting path for `cutadapt_se` and params for cutadapt wrapper.

* fix formatting

* better error message in gene2symbol.R

* apply dirty fix in biomart.yaml

the same fix as applied here:
https://github.com/snakemake-workflows/rna-seq-kallisto-sleuth/commit/5bf0c528e7e4cf1687ba72b95a547ad16a33130e

original solution can be found here:
https://stackoverflow.com/a/77370920

---------

Co-authored-by: David Laehnemann <david.laehnemann@hhu.de>
---
 workflow/envs/biomart.yaml     |  4 ++++
 workflow/rules/common.smk      |  4 +++-
 workflow/rules/trim.smk        |  2 +-
 workflow/scripts/gene2symbol.R | 11 ++++++++++-
 4 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/workflow/envs/biomart.yaml b/workflow/envs/biomart.yaml
index 04342c7..e9fe9c7 100644
--- a/workflow/envs/biomart.yaml
+++ b/workflow/envs/biomart.yaml
@@ -5,3 +5,7 @@ channels:
 dependencies:
   - bioconductor-biomart =2.56
   - r-tidyverse =2.0
+  # remove once we can update to bioconductor-biomart of the 3.18 release, which will
+  # include this proper fix for the underlying compatibility issue:
+  # https://github.com/Bioconductor/BiocFileCache/pull/50
+  - r-dbplyr=2.3.4
diff --git a/workflow/rules/common.smk b/workflow/rules/common.smk
index 515aef1..1817063 100644
--- a/workflow/rules/common.smk
+++ b/workflow/rules/common.smk
@@ -119,7 +119,9 @@ def get_fq(wildcards):
                 )
             )
         # single end sample
-        return {"fq1": "trimmed/{sample}_{unit}_single.fastq.gz".format(**wildcards)}
+        return {
+            "fq1": "results/trimmed/{sample}_{unit}_single.fastq.gz".format(**wildcards)
+        }
     else:
         # no trimming, use raw reads
         u = units.loc[(wildcards.sample, wildcards.unit)]
diff --git a/workflow/rules/trim.smk b/workflow/rules/trim.smk
index 9acf527..829e4d5 100644
--- a/workflow/rules/trim.smk
+++ b/workflow/rules/trim.smk
@@ -49,7 +49,7 @@ rule cutadapt_se:
         "logs/cutadapt/{sample}_{unit}.log",
     params:
         extra=config["params"]["cutadapt-se"],
-        adapters_r1=lambda w: str(units.loc[w.sample].loc[w.unit, "adapters"]),
+        adapters=lambda w: str(units.loc[w.sample].loc[w.unit, "adapters"]),
     threads: 8
     wrapper:
         "v1.21.4/bio/cutadapt/se"
diff --git a/workflow/scripts/gene2symbol.R b/workflow/scripts/gene2symbol.R
index c14ef85..c2abe01 100644
--- a/workflow/scripts/gene2symbol.R
+++ b/workflow/scripts/gene2symbol.R
@@ -1,5 +1,7 @@
 library(biomaRt)
 library(tidyverse)
+# useful error messages upon aborting
+library("cli")
 
 # this variable holds a mirror name until
 # useEnsembl succeeds ("www" is last, because 
@@ -24,7 +26,14 @@ while ( class(mart)[[1]] != "Mart" ) {
       # change or make configurable if you want more or
       # less rounds of tries of all the mirrors
       if (rounds >= 3) {
-        stop(
+        cli_abort(
+          str_c(
+            "Have tried all 4 available Ensembl biomaRt mirrors ",
+            rounds,
+            " times. You might have a connection problem, or no mirror is responsive.\n",
+            "The last error message was:\n",
+            message(e)
+          )
         )
       }
       # hop to next mirror