Merge pull request #46 from cokelaer/main

fix regression and update pyproject

Author: cokelaer

README.rst

@@ -177,6 +177,8 @@ Changelog
 ========= ====================================================================
 Version   Description
 ========= ====================================================================
+0.19.4    * Fix regression due to new sequana version
+          * 
 0.19.3    * fix regression with click to set the default rRNA to 'rRNA' again.
 0.19.2    * fix bowtie1 regression in the log file, paired end case in
             multiqc and rnadiff script (regression)

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "sequana-rnaseq"
-version = "0.19.3"
+version = "0.19.4"
 description = "A RNAseq pipeline from raw reads to feature counts"
 authors = ["Sequana Team"]
 license = "BSD-3"
@@ -34,9 +34,10 @@ packages = [
 
 [tool.poetry.dependencies]
 python = ">=3.8,<4.0"
-sequana = ">=0.16.0"
-sequana_pipetools = ">=0.16.2"
+sequana = ">=0.17.2"
+sequana_pipetools = ">=1.0.2"
 click-completion = "^0.5.2"
+pulp = "<2.8.0"
 
 
 [tool.poetry.scripts]

sequana_pipelines/rnaseq/config.yaml

@@ -310,12 +310,10 @@ feature_counts:
 # - threads: number of threads to be used
 bam_coverage:
     do: false
-    options:
-    binSize: 10
-    genomeSize: 2150570000  ##mm10
-    extendReads: 65
-    minFragmentLength: 0 #Note that a value other than 0 will exclude all single-end reads.
-    maxFragmentLength: 0 #A value of 0 disables filtering and is needed for including single-end and orphan reads.
+    options: "--binSize 10 --genomeSize 2150000000"
+    #extendReads: 65
+    #minFragmentLength: 0 #Note that a value other than 0 will exclude all single-end reads.
+    #maxFragmentLength: 0 #A value of 0 disables filtering and is needed for including single-end and orphan reads.
     threads: 4
     resources:
         mem: 20G

sequana_pipelines/rnaseq/main.py

@@ -265,7 +265,7 @@ def main(**options):
         gff = GFF3(gff_file)
         df_gff = gff.df  # This takes one minute on eukaryotes. No need to
         valid_features = gff.features  # about 3 seconds
-        valid_attributes = gff.attributes  # about 10 seconds
+        valid_attributes = gff.get_attributes()  # about 10 seconds
 
         # first check the rRNA feature
         if cfg["general"]["rRNA_feature"] and cfg["general"]["rRNA_feature"] not in valid_features:

sequana_pipelines/rnaseq/rnaseq.rules

@@ -366,7 +366,7 @@ elif manager.config.trimming.software_choice == "fastp":
         options_fastp += "--disable_quality_filtering"
 
     rule fastp:
-        input: 
+        input:
             sample=manager.getrawdata()
         output:
             trimmed=__clean_fastq__output,
@@ -425,7 +425,7 @@ else:
 # ==========  decompress and sanity check
 #
 if int(config['bowtie1_mapping_rna']['nreads']) != -1:
-    extra = int(config['bowtie1_mapping_rna']['nreads']) % 4
+    extra = int(config['bowtie1_mapping_rna']['nreads']) * 4
     config['bowtie1_mapping_rna']['nreads'] -= extra
 
 
@@ -723,7 +723,7 @@ else :
 fc_outdir = "post_analysis/feature_counts/"
 
 if manager.config.feature_counts.do and manager.config.general.aligner not in ['salmon']:
-    # Guessing strandness is not always straightfoward; Even when we set it; 
+    # Guessing strandness is not always straightfoward; Even when we set it;
     # collaborators may want to look at the other options. So, we compute
     # everything with the 3 different options of strandness.
     # We will copy one of them based on our criteria, but all 3 will be
@@ -769,7 +769,7 @@ if manager.config.feature_counts.do and manager.config.general.aligner not in ['
             f"{manager.wrappers}/wrappers/feature_counts"
 
 
-    # ===================== guessing the strand 
+    # ===================== guessing the strand
     #
     __guess_strandness__output = expand(fc_outdir + "{sample}_feature.out", sample=manager.samples)
     rule guess_strandness:
@@ -783,7 +783,7 @@ if manager.config.feature_counts.do and manager.config.general.aligner not in ['
             # We compute all strandness
             import sequana.featurecounts as fc
 
-            mfc = fc.MultiFeatureCount(rnaseq_folder=".", 
+            mfc = fc.MultiFeatureCount(rnaseq_folder=".",
                         tolerance=manager.config.feature_counts.tolerance)
             mfc.df.to_csv(output.summary)
 
@@ -943,8 +943,8 @@ if config["rnaseqc"]["do"] and config['general']['aligner'] != 'salmon':
         output:
             gtf = temp("tmp/test.gtf")
         run:
-            # If input GTF has no exon or genes, an error message is printed and 
-            # no files are created. This seems to be an issue in rnaseqc. 
+            # If input GTF has no exon or genes, an error message is printed and
+            # no files are created. This seems to be an issue in rnaseqc.
             # So, we create dummy gene and exon
             with open(output.gtf, "w") as ff:
                 ff.write(open(input['gtf'], "r").read())