Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Remove code duplication in main workflow #3

Merged
merged 19 commits into from
Oct 27, 2023
Merged

Remove code duplication in main workflow #3

Changes from 1 commit
Commits
Show all changes
19 commits
Select commit Hold shift + click to select a range
d127597
Remove code duplication in main workflow
Sep 21, 2023
a059ef2
Add missing curly brace
Sep 21, 2023
2c92d71
Use switch for setting species id
Sep 21, 2023
b52ec46
Add missing breaks to switch
Sep 21, 2023
3138b1c
Exclude nextflow cache dir and logs from version control
Sep 21, 2023
9dfe14d
Rename gene metadata files to match other file path structure
Sep 21, 2023
ddc800e
Rename MotMapsFile_gw and *_lb parameters to just MotifMapsFile
Sep 21, 2023
f8aa8cf
Refine gitignore
Sep 21, 2023
3b1acb2
Merge branch 'dev' into feature/remove-duplication-from-main-workflow
Sep 21, 2023
b414990
Ignore bin folder
Sep 21, 2023
8a7df76
Merge branch 'dev' into feature/remove-duplication-from-main-workflow
nicomaper Sep 27, 2023
c645c02
Merge branch 'feature/automated-e2e-test' into feature/remove-duplica…
Sep 28, 2023
b81569c
Revert accidental change of p-value for locus-based
Sep 28, 2023
99ea93b
Provide gene coords for gw mode only
Sep 28, 2023
b5637ee
Correct test input file paths due to rename
Sep 28, 2023
d70ee8d
Pass full params object from main workflow to gw and lb workflows
Sep 28, 2023
6baf449
Merge branch 'dev' into feature/remove-duplication-from-main-workflow
nicomaper Sep 29, 2023
bef0d07
update on .gitignore
nicomaper Sep 29, 2023
9200afb
Merge branch 'dev' into feature/remove-duplication-from-main-workflow
Oct 16, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
117 changes: 31 additions & 86 deletions mini_ac.nf
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,99 +7,44 @@ workflow MINIAC {

params.OBO_file = "$projectDir/data/ontologies/go.obo"

if (params.mode == "genome_wide" && params.species == "maize_v4") {

params.MotMapsFile_gw = "$projectDir/data/zma_v4/zma_v4_genome_wide_motif_mappings.bed"
params.Non_cod_genome = "$projectDir/data/zma_v4/zma_v4_noncod_merged.bed"
params.Faix_file = "$projectDir/data/zma_v4/zma_v4.fasta.fai"
params.Motif_tf_file = "$projectDir/data/zma_v4/zma_v4_motif_TF_file.txt"
params.Genes_coords = "$projectDir/data/zma_v4/zma_v4_genes_coords_sorted.bed"
params.Feature_file = "$projectDir/data/zma_v4/zma_v4_go_gene_file.txt"
params.TF_fam_file = "$projectDir/data/zma_v4/zma_v4_TF_family_file.txt"
params.Genes_metadata = "$projectDir/data/zma_v4/maize_v4_gene_metadata_file.txt"
params.P_val = 0.1
// define species id used for data subfolder and data file prefix
def species = "ath"
if(params.species == "maize_v4") {
species = "zma_v4"
} else if(params.species == "maize_v5") {
species = "zma_v5"
} else {
exit 1, "MINI-AC can only be run for the species 'arabidopsis', 'maize_v4' and 'maize_v5'. Instead it got '${params.species}'."
}

// set input data parameters shared between genome-wide and locus-based modes
params.Faix_file = "$projectDir/data/${species}/${species}.fasta.fai"
params.Motif_tf_file = "$projectDir/data/${species}/${species}_motif_TF_file.txt"
params.Genes_coords = "$projectDir/data/${species}/${species}_genes_coords_sorted.bed"
params.Feature_file = "$projectDir/data/${species}/${species}_go_gene_file.txt"
params.TF_fam_file = "$projectDir/data/${species}/${species}_TF_family_file.txt"
params.Genes_metadata = "$projectDir/data/${species}/${species}_gene_metadata_file.txt"
hdbeukel marked this conversation as resolved.
Show resolved Hide resolved

// set defaut p-value
params.P_val = 0.1
hdbeukel marked this conversation as resolved.
Show resolved Hide resolved

if (params.mode == "genome_wide") {

genome_wide_miniac(params.OutDir, params.ACR_dir, params.Filter_set_genes, params.Set_genes_dir, params.One_filtering_set, params.DE_genes, params.DE_genes_dir, params.One_DE_set, params.P_val, params.Bps_intersect, params.Second_gene_annot, params.Second_gene_dist, params.MotMapsFile_gw, params.Non_cod_genome, params.Faix_file, params.Motif_tf_file, params.Genes_coords, params.Feature_file, params.OBO_file, params.TF_fam_file, params.Genes_metadata)
}

else if (params.mode == "genome_wide" && params.species == "maize_v5") {
params.MotMapsFile_gw = "$projectDir/data/${species}/${species}_genome_wide_motif_mappings.bed"
hdbeukel marked this conversation as resolved.
Show resolved Hide resolved
params.Non_cod_genome = "$projectDir/data/${species}/${species}_noncod_merged.bed"

params.MotMapsFile_gw = "$projectDir/data/zma_v5/zma_v5_genome_wide_motif_mappings.bed"
params.Non_cod_genome = "$projectDir/data/zma_v5/zma_v5_noncod_merged.bed"
params.Faix_file = "$projectDir/data/zma_v5/zma_v5.fasta.fai"
params.Motif_tf_file = "$projectDir/data/zma_v5/zma_v5_motif_TF_file.txt"
params.Genes_coords = "$projectDir/data/zma_v5/zma_v5_genes_coords_sorted.bed"
params.Feature_file = "$projectDir/data/zma_v5/zma_v5_go_gene_file.txt"
params.TF_fam_file = "$projectDir/data/zma_v5/zma_v5_TF_family_file.txt"
params.Genes_metadata = "$projectDir/data/zma_v5/maize_v5_gene_metadata_file.txt"
params.P_val = 0.1

genome_wide_miniac(params.OutDir, params.ACR_dir, params.Filter_set_genes, params.Set_genes_dir, params.One_filtering_set, params.DE_genes, params.DE_genes_dir, params.One_DE_set, params.P_val, params.Bps_intersect, params.Second_gene_annot, params.Second_gene_dist, params.MotMapsFile_gw, params.Non_cod_genome, params.Faix_file, params.Motif_tf_file, params.Genes_coords, params.Feature_file, params.OBO_file, params.TF_fam_file, params.Genes_metadata)
}

else if (params.mode == "genome_wide" && params.species == "arabidopsis") {

params.MotMapsFile_gw = "$projectDir/data/ath/ath_genome_wide_motif_mappings.bed"
params.Non_cod_genome = "$projectDir/data/ath/ath_noncod_merged.bed"
params.Faix_file = "$projectDir/data/ath/ath.fasta.fai"
params.Motif_tf_file = "$projectDir/data/ath/ath_motif_TF_file.txt"
params.Genes_coords = "$projectDir/data/ath/ath_genes_coords_sorted.bed"
params.Feature_file = "$projectDir/data/ath/ath_go_gene_file.txt"
params.TF_fam_file = "$projectDir/data/ath/ath_TF_family_file.txt"
params.Genes_metadata = "$projectDir/data/ath/arabidopsis_gene_metadata_file.txt"
params.P_val = 0.1

genome_wide_miniac(params.OutDir, params.ACR_dir, params.Filter_set_genes, params.Set_genes_dir, params.One_filtering_set, params.DE_genes, params.DE_genes_dir, params.One_DE_set, params.P_val, params.Bps_intersect, params.Second_gene_annot, params.Second_gene_dist, params.MotMapsFile_gw, params.Non_cod_genome, params.Faix_file, params.Motif_tf_file, params.Genes_coords, params.Feature_file, params.OBO_file, params.TF_fam_file, params.Genes_metadata)

}

else if (params.mode == "locus_based" && params.species == "maize_v4") {

else if (params.mode == "locus_based") {

params.MotMapsFile_lb = "$projectDir/data/zma_v4/zma_v4_locus_based_motif_mappings_5kbup_1kbdown.bed"
params.Promoter_file = "$projectDir/data/zma_v4/zma_v4_promoter_5kbup_1kbdown_sorted.bed"
params.Faix_file = "$projectDir/data/zma_v4/zma_v4.fasta.fai"
params.Motif_tf_file = "$projectDir/data/zma_v4/zma_v4_motif_TF_file.txt"
params.Feature_file = "$projectDir/data/zma_v4/zma_v4_go_gene_file.txt"
params.TF_fam_file = "$projectDir/data/zma_v4/zma_v4_TF_family_file.txt"
params.Genes_metadata = "$projectDir/data/zma_v4/maize_v4_gene_metadata_file.txt"
params.P_val = 0.01
params.MotMapsFile_lb = "$projectDir/data/${species}/${species}_locus_based_motif_mappings_5kbup_1kbdown.bed"
params.Promoter_file = "$projectDir/data/${species}/${species}_promoter_5kbup_1kbdown_sorted.bed"

locus_based_miniac(params.OutDir, params.ACR_dir, params.Filter_set_genes, params.Set_genes_dir, params.One_filtering_set, params.DE_genes, params.DE_genes_dir, params.One_DE_set, params.P_val, params.Bps_intersect, params.MotMapsFile_lb, params.Promoter_file, params.Faix_file, params.Motif_tf_file, params.Feature_file, params.OBO_file, params.TF_fam_file, params.Genes_metadata)

}

else if (params.mode == "locus_based" && params.species == "maize_v5") {

params.MotMapsFile_lb = "$projectDir/data/zma_v5/zma_v5_locus_based_motif_mappings_5kbup_1kbdown.bed"
params.Promoter_file = "$projectDir/data/zma_v5/zma_v5_promoter_5kbup_1kbdown_sorted.bed"
params.Faix_file = "$projectDir/data/zma_v5/zma_v5.fasta.fai"
params.Motif_tf_file = "$projectDir/data/zma_v5/zma_v5_motif_TF_file.txt"
params.Feature_file = "$projectDir/data/zma_v5/zma_v5_go_gene_file.txt"
params.TF_fam_file = "$projectDir/data/zma_v5/zma_v5_TF_family_file.txt"
params.Genes_metadata = "$projectDir/data/zma_v5/maize_v5_gene_metadata_file.txt"
params.P_val = 0.01

locus_based_miniac(params.OutDir, params.ACR_dir, params.Filter_set_genes, params.Set_genes_dir, params.One_filtering_set, params.DE_genes, params.DE_genes_dir, params.One_DE_set, params.P_val, params.Bps_intersect, params.MotMapsFile_lb, params.Promoter_file, params.Faix_file, params.Motif_tf_file, params.Feature_file, params.OBO_file, params.TF_fam_file, params.Genes_metadata)


} else {
exit 1, "MINI-AC can only be run using the modes 'genome_wide' or 'locus_based'. Instead it got '${params.mode}'."
}

else if (params.mode == "locus_based" && params.species == "arabidopsis") {

params.MotMapsFile_lb = "$projectDir/data/ath/ath_locus_based_motif_mappings_5kbup_1kbdown.bed"
params.Promoter_file = "$projectDir/data/ath/ath_promoter_5kbup_1kbdown_sorted.bed"
params.Faix_file = "$projectDir/data/ath/ath.fasta.fai"
params.Motif_tf_file = "$projectDir/data/ath/ath_motif_TF_file.txt"
params.Feature_file = "$projectDir/data/ath/ath_go_gene_file.txt"
params.TF_fam_file = "$projectDir/data/ath/ath_TF_family_file.txt"
params.Genes_metadata = "$projectDir/data/ath/arabidopsis_gene_metadata_file.txt"
params.P_val = 0.01

locus_based_miniac(params.OutDir, params.ACR_dir, params.Filter_set_genes, params.Set_genes_dir, params.One_filtering_set, params.DE_genes, params.DE_genes_dir, params.One_DE_set, params.P_val, params.Bps_intersect, params.MotMapsFile_lb, params.Promoter_file, params.Faix_file, params.Motif_tf_file, params.Feature_file, params.OBO_file, params.TF_fam_file, params.Genes_metadata)
}

else {
exit 1, "MINI-AC can only be run using the modes 'genome_wide' and 'locus_based', and with the species 'arabidopsis', 'maize_v4' and 'maize_v5'. Instead it got '${params.species}' and '${params.mode}' "
}
}


Expand Down