From e4c8ebf06bf267002d16d56e3772cfe1d6845e72 Mon Sep 17 00:00:00 2001
From: Valentina Galata <valentina.galata@uni.lu>
Date: Sun, 16 Aug 2020 19:03:10 +0200
Subject: [PATCH] updated tax rules: kaiju for all proteins, kraken2 for
filtered contigs only
---
workflow/rules/taxonomy.smk | 64 +++++++++++--------------------------
workflow/steps/taxonomy.smk | 29 ++++-------------
2 files changed, 25 insertions(+), 68 deletions(-)
diff --git a/workflow/rules/taxonomy.smk b/workflow/rules/taxonomy.smk
index 87b7e12..b21c3d4 100644
--- a/workflow/rules/taxonomy.smk
+++ b/workflow/rules/taxonomy.smk
@@ -5,17 +5,16 @@
rule tax_kraken2_contigs:
input:
- contigs=os.path.join(RESULTS_DIR, "assembly/{rtype}/{tool}/{atype}.fasta"),
+ contigs=os.path.join(RESULTS_DIR, "assembly/{rtype}/{tool}/ASSEMBLY.FILTERED.fasta"),
db=lambda wildcards: config["kraken2"]["db"][wildcards.db]
output:
- labels=os.path.join(RESULTS_DIR, "taxonomy/kraken2/{rtype}.{tool}.{atype}.{db}.labels.txt"),
- report=os.path.join(RESULTS_DIR, "taxonomy/kraken2/{rtype}.{tool}.{atype}.{db}.report.txt")
+ labels=os.path.join(RESULTS_DIR, "taxonomy/kraken2/{rtype}.{tool}.{db}.labels.txt"),
+ report=os.path.join(RESULTS_DIR, "taxonomy/kraken2/{rtype}.{tool}.{db}.report.txt")
log:
- "logs/kraken2.{rtype}.{tool}.{atype}.{db}.log"
+ "logs/kraken2.{rtype}.{tool}.{db}.log"
wildcard_constraints:
rtype="|".join(READ_TYPES),
tool="|".join(ASSEMBLERS),
- atype="ASSEMBLY|ASSEMBLY.FILTERED",
db="|".join(config["kraken2"]["db"].keys())
threads:
config["kraken2"]["threads"]
@@ -85,68 +84,43 @@ rule tax_kraken2_lr:
##################################################
# Kaiju
-rule tax_kaiju_cdhit:
+rule tax_kaiju:
input:
- faa=os.path.join(RESULTS_DIR, "analysis/cdhit/{rtype1}_{tool1}__{rtype2}_{tool2}.faa"),
+ faa=os.path.join(RESULTS_DIR, "annotation/prodigal/{rtype}/{tool}/proteins.faa"),
nodes=lambda wildcards: os.path.join(config["kaiju"]["db"][wildcards.db], "nodes.dmp"),
names=lambda wildcards: os.path.join(config["kaiju"]["db"][wildcards.db], "names.dmp"),
fmi=lambda wildcards: os.path.join(config["kaiju"]["db"][wildcards.db], "%s.fmi" % wildcards.db),
output:
- out=temp(os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{rtype1}_{tool1}__{rtype2}_{tool2}.{db}.tsv.tmp")),
- names=os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{rtype1}_{tool1}__{rtype2}_{tool2}.{db}.tsv")
+ out=temp(os.path.join(RESULTS_DIR, "taxonomy/kaiju/{rtype}.{tool}.{db}.tsv.tmp")),
+ names=os.path.join(RESULTS_DIR, "taxonomy/kaiju/{rtype}.{tool}.{db}.tsv")
log:
- "logs/kaiju.cdhit.{rtype1}_{tool1}__{rtype2}_{tool2}.{db}.log"
+ "logs/kaiju.{rtype}.{tool}.{db}.log"
wildcard_constraints:
- rtype1="|".join(READ_TYPES),
- rtype2="|".join(READ_TYPES),
- tool1="|".join(ASSEMBLERS),
- tool2="|".join(ASSEMBLERS),
+ rtype="|".join(READ_TYPES),
+ tool="|".join(ASSEMBLERS),
db="|".join(config["kaiju"]["db"].keys())
threads:
config["kaiju"]["threads"]
conda:
"../envs/kaiju.yaml"
message:
- "Tax. classification w/ Kaiju ({wildcards.db}, CD-HIT)"
+ "Tax. classification w/ Kaiju ({wildcards.db}, proteins)"
shell:
"(date && "
"kaiju -t {input.nodes} -f {input.fmi} -i {input.faa} -o {output.out} -z {threads} -v -p && "
"kaiju-addTaxonNames -p -t {input.nodes} -n {input.names} -i {output.out} -o {output.names} && "
"date) &> {log}"
-# rule tax_kaiju_summary:
-# input:
-# out=os.path.join(RESULTS_DIR, "taxonomy/kaiju/{bname}.{db}.tsv"),
-# nodes=lambda wildcards: os.path.join(config["kaiju"]["db"][wildcards.db], "nodes.dmp"),
-# names=lambda wildcards: os.path.join(config["kaiju"]["db"][wildcards.db], "names.dmp"),
-# output:
-# os.path.join(RESULTS_DIR, "taxonomy/kaiju/{bname}.{db}.{rank}")
-# wildcard_constraints:
-# db="|".join(config["kaiju"]["db"].keys()),
-# rank="|".join(config["kaiju"]["ranks"])
-# threads:
-# 1
-# conda:
-# "../envs/kaiju.yaml"
-# message:
-# "Tax. classification summary w/ Kaiju ({wildcards.db}, {wildcards.rank})"
-# shell:
-# "kaiju2table -p -t {input.nodes} -n {input.names} -r {wildcards.rank} -o {output} {input.out} -v"
-
-rule tax_kaiju_cdhit_summary:
+rule tax_kaiju_summary:
input:
- out=expand(
- os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{combi}.{{db}}.tsv"),
- combi=["%s_%s__%s_%s" % (p[0][0], p[0][1], p[1][0], p[1][1]) for p in READ_ASSEMBLER_PAIRS]
- ) + expand(
- os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{combi}.{{db}}.tsv"),
- combi=["%s_%s__%s_%s" % (p[1][0], p[1][1], p[0][0], p[0][1]) for p in READ_ASSEMBLER_PAIRS]
- ),
+ out=os.path.join(RESULTS_DIR, "taxonomy/kaiju/{rtype}.{tool}.{db}.tsv"),
nodes=lambda wildcards: os.path.join(config["kaiju"]["db"][wildcards.db], "nodes.dmp"),
- names=lambda wildcards: os.path.join(config["kaiju"]["db"][wildcards.db], "names.dmp")
+ names=lambda wildcards: os.path.join(config["kaiju"]["db"][wildcards.db], "names.dmp"),
output:
- os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{db}.summary.{rank}.tsv")
+ os.path.join(RESULTS_DIR, "taxonomy/kaiju/{rtype}.{tool}.{db}.summary.{rank}.tsv")
wildcard_constraints:
+ rtype="|".join(READ_TYPES),
+ tool="|".join(ASSEMBLERS),
db="|".join(config["kaiju"]["db"].keys()),
rank="|".join(config["kaiju"]["ranks"])
threads:
@@ -156,4 +130,4 @@ rule tax_kaiju_cdhit_summary:
message:
"Tax. classification summary w/ Kaiju ({wildcards.db}, {wildcards.rank})"
shell:
- "kaiju2table -p -t {input.nodes} -n {input.names} -r {wildcards.rank} -o {output} {input.out} -v"
\ No newline at end of file
+ "kaiju2table -p -t {input.nodes} -n {input.names} -r {wildcards.rank} -o {output} {input.out} -v"
diff --git a/workflow/steps/taxonomy.smk b/workflow/steps/taxonomy.smk
index 008c989..eb39f66 100644
--- a/workflow/steps/taxonomy.smk
+++ b/workflow/steps/taxonomy.smk
@@ -7,9 +7,8 @@ rule TAXONOMY:
input:
# Kraken2
expand(
- os.path.join(RESULTS_DIR, "taxonomy/kraken2/{rtype_tool}.{atype}.{db}.{otype}.txt"),
+ os.path.join(RESULTS_DIR, "taxonomy/kraken2/{rtype_tool}.{db}.{otype}.txt"),
rtype_tool=["%s.%s" % (rtype, tool) for rtype, tool in READ_ASSEMBLERS],
- atype=["ASSEMBLY", "ASSEMBLY.FILTERED"],
db=config["kraken2"]["db"].keys(),
otype=["labels", "report"]
) if "kraken2" in config["steps_taxonomy"] else [],
@@ -26,31 +25,15 @@ rule TAXONOMY:
) if "kraken2" in config["steps_taxonomy"] else [],
# Kaiju
expand(
- os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{combi}.{db}.tsv"),
- combi=["%s_%s__%s_%s" % (p[0][0], p[0][1], p[1][0], p[1][1]) for p in READ_ASSEMBLER_PAIRS],
- db=config["kaiju"]["db"].keys()
- ) if "kaiju" in config["steps_taxonomy"] else [],
- expand(
- os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{combi}.{db}.tsv"),
- combi=["%s_%s__%s_%s" % (p[1][0], p[1][1], p[0][0], p[0][1]) for p in READ_ASSEMBLER_PAIRS],
+ os.path.join(RESULTS_DIR, "taxonomy/kaiju/{rtype_tool}.{db}.tsv"),
+ rtype_tool=["%s.%s" % (rtype, tool) for rtype, tool in READ_ASSEMBLERS],
db=config["kaiju"]["db"].keys()
) if "kaiju" in config["steps_taxonomy"] else [],
expand(
- os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{db}.summary.{rank}.tsv"),
+ os.path.join(RESULTS_DIR, "taxonomy/kaiju/{rtype_tool}.{db}.summary.{rank}.tsv"),
+ rtype_tool=["%s.%s" % (rtype, tool) for rtype, tool in READ_ASSEMBLERS],
db=config["kaiju"]["db"].keys(),
rank=config["kaiju"]["ranks"]
- ) if "kaiju" in config["steps_taxonomy"] else []
- # expand(
- # os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{combi}.{db}.{rank}"),
- # combi=["%s_%s__%s_%s" % (p[0][0], p[0][1], p[1][0], p[1][1]) for p in READ_ASSEMBLER_PAIRS],
- # db=config["kaiju"]["db"].keys(),
- # rank=config["kaiju"]["ranks"]
- # ) if "kaiju" in config["steps_taxonomy"] else [],
- # expand(
- # os.path.join(RESULTS_DIR, "taxonomy/kaiju/cdhit.{combi}.{db}.{rank}"),
- # combi=["%s_%s__%s_%s" % (p[1][0], p[1][1], p[0][0], p[0][1]) for p in READ_ASSEMBLER_PAIRS],
- # db=config["kaiju"]["db"].keys(),
- # rank=config["kaiju"]["ranks"]
- # ) if "kaiju" in config["steps_taxonomy"] else [],
+ ) if "kaiju" in config["steps_taxonomy"] else [],
output:
touch("status/taxonomy.done")
--
GitLab