Merge pull request #103 from gbouras13/dev

V0.10.0 flye extra parama

HISTORY.md

@@ -5,6 +5,7 @@
 * Updates Medaka to v2.0.1, implementing the `--bacteria` option by default.
 * This is based on the recommendations of Ryan Wick [here](https://rrwick.github.io/2024/10/17/medaka-v2.html) who found it improved assemblies due to (likely) enhanced methylation error correction.
 * If you still want to specify a Medaka model, the flag `--medaka_override` has been added. You need to include this along with your model via `--medakaModel`. This is most likely useful for older R9 data.
+* 
 
 ## v0.9.1 Updates (8 October 2024)
 

hybracter/__main__.py

@@ -162,6 +162,12 @@ def common_options(func):
             is_flag=True,
             default=False,
         ),
+        click.option(
+            "--extra_params_flye",
+            help="Use this if want to add extra parameters to Flye.",
+            type=str,
+            default=None
+        ),
 
         click.option(
             "--use-conda/--no-use-conda",
@@ -406,6 +412,7 @@ def hybrid(
     depth_filter,
     mac,
     medaka_override,
+    extra_params_flye,
     log,
     configfile,
     **kwargs
@@ -433,6 +440,7 @@ def hybrid(
             "auto": auto,
             "mac": mac,
             "medaka_override": medaka_override,
+            "extra_params_flye": extra_params_flye,
             "depth_filter": depth_filter,
             "single": False,
             "logic": logic,
@@ -532,6 +540,7 @@ def hybrid_single(
     auto,
     mac,
     medaka_override,
+    extra_params_flye,
     logic,
     depth_filter,
     log,
@@ -564,6 +573,7 @@ def hybrid_single(
             "auto": auto,
             "mac": mac,
             "medaka_override": medaka_override,
+            "extra_params_flye": extra_params_flye,
             "depth_filter": depth_filter,
             "single": True,
             "logic": logic,
@@ -627,6 +637,7 @@ def long(
     auto,
     mac,
     medaka_override,
+    extra_params_flye,
     logic,
     depth_filter,
     log,
@@ -655,6 +666,7 @@ def long(
             "auto": auto,
             "mac": mac,
             "medaka_override": medaka_override,
+            "extra_params_flye": extra_params_flye,
             "depth_filter": depth_filter,
             "single": False,
             "logic": logic,
@@ -732,6 +744,7 @@ def long_single(
     auto,
     mac,
     medaka_override,
+    extra_params_flye,
     logic,
     depth_filter,
     configfile,
@@ -760,6 +773,7 @@ def long_single(
             "auto": auto,
             "mac": mac,
             "medaka_override": medaka_override,
+            "extra_params_flye": extra_params_flye,
             "depth_filter": depth_filter,
             "single": True,
             "logic": logic,
@@ -913,6 +927,7 @@ def test_hybrid(
     no_pypolca,
     mac,
     medaka_override,
+    extra_params_flye,
     contaminants,
     dnaapler_custom_db,
     logic,
@@ -941,6 +956,7 @@ def test_hybrid(
             "auto": auto,
             "mac": mac,
             "medaka_override": medaka_override,
+            "extra_params_flye": extra_params_flye,
             "logic": logic,
             "depth_filter": depth_filter,
             "configfile": configfile
@@ -998,6 +1014,7 @@ def test_long(
     auto,
     mac,
     medaka_override,
+    extra_params_flye,
     logic,
     depth_filter,
     configfile,
@@ -1023,6 +1040,7 @@ def test_long(
             "auto": auto,
             "mac": mac,
             "medaka_override": medaka_override,
+            "extra_params_flye": extra_params_flye,
             "logic": logic,
             "depth_filter": depth_filter,
             "configfile": configfile

hybracter/workflow/hybrid.smk

@@ -70,6 +70,14 @@ MIN_DEPTH = config.args.min_depth
 AUTO = config.args.auto
 MAC = config.args.mac
 MEDAKA_OVERRIDE = config.args.medaka_override
+EXTRA_PARAMS_FLYE = config.args.extra_params_flye
+# flag to add extra arguments to flye
+ADD_TO_FLYE =  False
+
+# Only if user specifies - otherwise None
+if EXTRA_PARAMS_FLYE:
+    ADD_TO_FLYE =  True
+    print(f"As you have used --extra_params_flye the extra parameters {EXTRA_PARAMS_FLYE} will be used with Flye.")
 
 # By default, hybracter (linux) will use --bacteria from v0.10.0 onwards 
 # To take advantage of improvements with medaka v2 https://rrwick.github.io/2024/10/17/medaka-v2.html

hybracter/workflow/long.smk

@@ -68,6 +68,15 @@ MIN_DEPTH = config.args.min_depth
 AUTO = config.args.auto
 MAC = config.args.mac
 MEDAKA_OVERRIDE = config.args.medaka_override
+EXTRA_PARAMS_FLYE = config.args.extra_params_flye
+# flag to add extra arguments to flye
+ADD_TO_FLYE =  False
+
+# Only if user specifies - otherwise None
+if EXTRA_PARAMS_FLYE:
+    ADD_TO_FLYE =  True
+    print(f"The extra parameters {EXTRA_PARAMS_FLYE} will be used with Flye ")
+
 
 # By default, hybracter (linux) will use --bacteria from v0.10.0 onwards 
 # To take advantage of improvements with medaka v2 https://rrwick.github.io/2024/10/17/medaka-v2.html

hybracter/workflow/rules/assembly/assemble.smk

@@ -20,15 +20,21 @@ rule assemble:
     params:
         model=FLYE_MODEL,
         dir=os.path.join(dir.out.assemblies, "{sample}"),
+        add_to_flye = ADD_TO_FLYE,
+        extra_params_flye = EXTRA_PARAMS_FLYE
     threads: config.resources.big.cpu
     benchmark:
         os.path.join(dir.out.bench, "assemble", "{sample}.txt")
     log:
         os.path.join(dir.out.stderr, "assemble", "{sample}.log"),
     shell:
         """
-        flye {params.model} {input.fastq} -t {threads}  --out-dir {params.dir} 2> {log}
-        flye --version > {output.version}
+        if [ "{params.add_to_flye}" = "True" ]; then 
+            flye {params.model} {input.fastq} -t {threads}  --out-dir {params.dir} {params.extra_params_flye} 2> {log}
+        else
+            flye {params.model} {input.fastq} -t {threads}  --out-dir {params.dir}  2> {log}
+        fi
+        flye --version > {output.version}   
         cp {output.info} {output.infocopy}
         """
 

hybracter/workflow/test_hybrid.smk

@@ -67,6 +67,15 @@ MIN_DEPTH = config.args.min_depth
 AUTO = config.args.auto
 MAC = config.args.mac
 MEDAKA_OVERRIDE = config.args.medaka_override
+EXTRA_PARAMS_FLYE = config.args.extra_params_flye
+# flag to add extra arguments to flye
+ADD_TO_FLYE =  False
+
+# Only if user specifies - otherwise None
+if EXTRA_PARAMS_FLYE:
+    ADD_TO_FLYE =  True
+    print(f"The extra parameters {EXTRA_PARAMS_FLYE} will be used with Flye ")
+
 
 # By default, hybracter (linux) will use --bacteria from v0.10.0 onwards 
 # To take advantage of improvements with medaka v2 https://rrwick.github.io/2024/10/17/medaka-v2.html

hybracter/workflow/test_long.smk

@@ -67,6 +67,14 @@ MIN_DEPTH = config.args.min_depth
 AUTO = config.args.auto
 MAC = config.args.mac
 MEDAKA_OVERRIDE = config.args.medaka_override
+EXTRA_PARAMS_FLYE = config.args.extra_params_flye
+# flag to add extra arguments to flye
+ADD_TO_FLYE =  False
+
+# Only if user specifies - otherwise None
+if EXTRA_PARAMS_FLYE:
+    ADD_TO_FLYE =  True
+    print(f"The extra parameters {EXTRA_PARAMS_FLYE} will be used with Flye ")
 
 # By default, hybracter (linux) will use --bacteria from v0.10.0 onwards 
 # To take advantage of improvements with medaka v2 https://rrwick.github.io/2024/10/17/medaka-v2.html

tests/test_hybracter.py

@@ -341,7 +341,7 @@ def test_hybracter_long(run_mac):
 # test override
 # bit useless for mac (as --bacteria won't work) but won't break anything so just leave
 
-def test_hybracter_long(run_mac):
+def test_hybracter_long_medaka_override(run_mac):
     """test hybracter long"""
     outdir: Path = "test_hybracter_output_medaka_override"
     input_csv: Path = test_data_path / "test_long_input.csv"
@@ -351,6 +351,19 @@ def test_hybracter_long(run_mac):
     exec_command(cmd)
     remove_directory(outdir)
 
+# test --extra_params_flye
+
+def test_hybracter_long_medaka_override(run_mac):
+    """test hybracter long"""
+    outdir: Path = "test_hybracter_output_medaka_override"
+    input_csv: Path = test_data_path / "test_long_input.csv"
+    cmd = f"hybracter long --input {input_csv} --threads {threads} --output {outdir} --databases {db_dir} --extra_params_flye \"--meta\" "
+    if run_mac:
+        cmd += " --mac"
+    exec_command(cmd)
+    remove_directory(outdir)
+
+
 def test_hybracter_long_auto(run_mac):
     """