Skip to content
Snippets Groups Projects
Commit 7bc8732e authored by Valentina Galata's avatar Valentina Galata
Browse files

analysis step update: updated quast/bbmap/cdhit conda YAML; other minor updates

parent 3a97208b
No related branches found
No related tags found
1 merge request!76Merge "cleanup" branch with "master" branch
Hide whitespace changes
Inline Side-by-side
config/config.yaml
+ 10
− 2
View file @ 7bc8732e
......@@ -2,12 +2,15 @@
# STEPS
# Pipeline steps
# steps: ["preprocessing", "assembly", "mapping", "annotation", "binning", "taxonomy", "analysis"]
steps: ["preprocessing", "assembly", "mapping", "annotation"]
# NOTE: no binning and taxonomic analysis
# steps: ["preprocessing", "assembly", "mapping", "annotation", "analysis"]
steps: ["preprocessing", "assembly", "mapping", "annotation", "analysis"]
# NOTE: currently not used
# Annotation sub-steps
# annotation_steps: ["plasmids", "crispr", "amr"]
# NOTE: currently not used
# Analysis sub-steps
# analysis_steps: ["quast", "prodigal", "cdhit", "mmseqs2"]
......@@ -198,6 +201,11 @@ mmseqs2:
# rbh: "/home/users/sbusi/apps/mmseqs/bin/mmseqs rbh"
# convertalis: "/home/users/sbusi/apps/mmseqs/bin/mmseqs convertalis"
# Seq. alignment
# https://mummer4.github.io/
mummer:
archive: "https://github.com/mummer4/mummer/releases/download/v3.9.4alpha/mummer-3.9.4alpha.tar.gz"
##############################
# Taxonomy
......
workflow/Snakefile
+ 0
− 12
View file @ 7bc8732e
......@@ -127,18 +127,6 @@ if "analysis" in STEPS:
"status/analysis.done"
]
# # Binning
# if 'binning' in STEPS:
# include:
# "steps/binning.smk"
# TARGETS += ["binning_for_ont.done"]
# # Taxonomy
# if 'taxonomy' in STEPS:
# include:
# "steps/taxonomy.smk"
# TARGETS += ["taxonomy_for_ont.done"]
# No targets
if len(TARGETS) == 0:
raise Exception('You are not serious. Nothing to be done? Really?')
......
workflow/envs/bbmap.yaml
+ 44
− 44
View file @ 7bc8732e
name: bbmap
channels:
- imp
- bioconda
- conda-forge
- bioconda
- defaults
dependencies:
- _libgcc_mutex=0.1
- _openmp_mutex=4.5
- alsa-lib=1.1.5
- bbmap=38.79
- fontconfig=2.13.1
- freetype=2.10.2
- giflib=5.2.1
- icu=67.1
- jpeg=9c
- lcms2=2.9
- libgcc-ng=9.2.0
- libgomp=9.2.0
- libiconv=1.15
- libpng=1.6.37
- libstdcxx-ng=9.2.0
- libtiff=4.1.0
- libuuid=2.32.1
- libwebp-base=1.1.0
- libxcb=1.13
- libxml2=2.9.10
- lz4-c=1.9.2
- openjdk=11.0.1
- pthread-stubs=0.4
- xorg-fixesproto=5.0
- xorg-inputproto=2.3.2
- xorg-kbproto=1.0.7
- xorg-libx11=1.6.9
- xorg-libxau=1.0.9
- xorg-libxdmcp=1.1.3
- xorg-libxext=1.3.4
- xorg-libxfixes=5.0.3
- xorg-libxi=1.7.10
- xorg-libxrender=0.9.10
- xorg-libxtst=1.2.3
- xorg-recordproto=1.14.2
- xorg-renderproto=0.11.1
- xorg-xextproto=7.3.0
- xorg-xproto=7.0.31
- xz=5.2.5
- zlib=1.2.11
- zstd=1.4.4
prefix: /home/users/sbusi/apps/miniconda3/envs/bbmap
- _libgcc_mutex=0.1=conda_forge
- _openmp_mutex=4.5=0_gnu
- alsa-lib=1.1.5=h516909a_1002
- bbmap=38.86=h1296035_0
- bzip2=1.0.8=h516909a_2
- fontconfig=2.13.1=h1056068_1002
- freetype=2.10.2=he06d7ca_0
- giflib=5.2.1=h516909a_2
- icu=67.1=he1b5a44_0
- jpeg=9d=h516909a_0
- lcms2=2.11=hbd6801e_0
- libgcc-ng=9.2.0=h24d8f2e_2
- libgomp=9.2.0=h24d8f2e_2
- libiconv=1.15=h516909a_1006
- libpng=1.6.37=hed695b0_1
- libstdcxx-ng=9.2.0=hdf63c60_2
- libtiff=4.1.0=hc7e4089_6
- libuuid=2.32.1=h14c3975_1000
- libwebp-base=1.1.0=h516909a_3
- libxcb=1.13=h14c3975_1002
- libxml2=2.9.10=h72b56ed_1
- lz4-c=1.9.2=he1b5a44_1
- openjdk=11.0.1=h600c080_1018
- pbzip2=1.1.13=0
- pthread-stubs=0.4=h14c3975_1001
- xorg-fixesproto=5.0=h14c3975_1002
- xorg-inputproto=2.3.2=h14c3975_1002
- xorg-kbproto=1.0.7=h14c3975_1002
- xorg-libx11=1.6.9=h516909a_0
- xorg-libxau=1.0.9=h14c3975_0
- xorg-libxdmcp=1.1.3=h516909a_0
- xorg-libxext=1.3.4=h516909a_0
- xorg-libxfixes=5.0.3=h516909a_1004
- xorg-libxi=1.7.10=h516909a_0
- xorg-libxrender=0.9.10=h516909a_1002
- xorg-libxtst=1.2.3=h516909a_1002
- xorg-recordproto=1.14.2=h516909a_1002
- xorg-renderproto=0.11.1=h14c3975_1002
- xorg-xextproto=7.3.0=h14c3975_1002
- xorg-xproto=7.0.31=h14c3975_1007
- xz=5.2.5=h516909a_0
- zlib=1.2.11=h516909a_1006
- zstd=1.4.4=h6597ccf_3
workflow/envs/cdhit.yaml
+ 1
− 1
View file @ 7bc8732e
channels:
- conda-forge
- bioconda
- conda-forge
- defaults
dependencies:
- _libgcc_mutex=0.1=conda_forge
......
workflow/envs/quast.yaml
+ 131
− 90
View file @ 7bc8732e
name: quast
channels:
- bioconda
- conda-forge
- bioconda
- defaults
dependencies:
- blast=2.6.0
- ca-certificates=2020.4.5.1
- certifi=2020.4.5.1
- circos=0.69.8
- cycler=0.10.0
- expat=2.2.9
- fontconfig=2.13.1
- freetype=2.10.2
- giflib=5.2.1
- glimmerhmm=3.0.4
- icu=64.2
- joblib=0.14.1
- jpeg=9c
- kiwisolver=1.2.0
- libblas=3.8.0
- libcblas=3.8.0
- libcxx=10.0.0
- libffi=3.2.1
- libgd=2.2.5
- libgfortran=4.0.0
- libiconv=1.15
- liblapack=3.8.0
- libopenblas=0.3.9
- libpng=1.6.37
- libtiff=4.1.0
- libwebp=1.0.2
- libxml2=2.9.10
- llvm-openmp=10.0.0
- lz4-c=1.9.2
- matplotlib=3.2.1
- matplotlib-base=3.2.1
- ncurses=6.1
- numpy=1.18.4
- openjdk=11.0.1
- openssl=1.1.1g
- perl=5.26.2
- perl-autoloader=5.74
- perl-carp=1.38
- perl-clone=0.42
- perl-config-general=2.63
- perl-digest-perl-md5=1.9
- perl-dynaloader=1.25
- perl-exporter=5.72
- perl-exporter-tiny=1.002001
- perl-extutils-makemaker=7.36
- perl-font-ttf=1.06
- perl-gd=2.71
- perl-io-string=1.08
- perl-list-moreutils=0.428
- perl-list-moreutils-xs=0.428
- perl-math-bezier=0.01
- perl-math-round=0.07
- perl-math-vecstat=0.08
- perl-module-implementation=0.09
- perl-module-runtime=0.016
- perl-number-format=1.75
- perl-params-validate=1.29
- perl-pathtools=3.75
- perl-readonly=2.05
- perl-regexp-common=2017060201
- perl-scalar-list-utils=1.52
- perl-set-intspan=1.19
- perl-statistics-basic=1.6611
- perl-svg=2.84
- perl-text-format=0.59
- perl-threaded=5.26.0
- perl-time-hires=1.9760
- perl-try-tiny=0.30
- perl-xml-parser=2.44
- perl-xsloader=0.24
- pip=20.1
- pyparsing=2.4.7
- python=3.6.10
- python-dateutil=2.8.1
- python_abi=3.6
- quast=5.0.2
- readline=8.0
- setuptools=46.3.0
- simplejson=3.8.1
- six=1.14.0
- sqlite=3.30.1
- tk=8.6.10
- tornado=6.0.4
- wheel=0.34.2
- xz=5.2.5
- zlib=1.2.11
- zstd=1.4.4
prefix: /home/users/sbusi/apps/miniconda3/envs/quast
- _libgcc_mutex=0.1=conda_forge
- _openmp_mutex=4.5=1_llvm
- alsa-lib=1.1.5=h516909a_1002
- blast=2.5.0=hc0b0e79_3
- boost=1.73.0=py36h6957fcd_0
- boost-cpp=1.73.0=h6c875cf_0
- bzip2=1.0.8=h516909a_2
- ca-certificates=2020.6.20=hecda079_0
- certifi=2020.6.20=py36h9f0ad1d_0
- circos=0.69.8=0
- cycler=0.10.0=py_2
- dbus=1.13.6=he372182_0
- expat=2.2.9=he1b5a44_2
- fontconfig=2.13.1=h86ecdb6_1001
- freetype=2.10.2=he06d7ca_0
- gettext=0.19.8.1=hc5be6a0_1002
- giflib=5.2.1=h516909a_2
- glib=2.65.0=h6f030ca_0
- glimmerhmm=3.0.4=pl526hf484d3e_3
- gst-plugins-base=1.14.5=h0935bb2_2
- gstreamer=1.14.5=h36ae1b5_2
- icu=64.2=he1b5a44_1
- joblib=0.15.1=py_0
- jpeg=9d=h516909a_0
- kiwisolver=1.2.0=py36hdb11119_0
- lcms2=2.11=hbd6801e_0
- ld_impl_linux-64=2.34=h53a641e_5
- libblas=3.8.0=17_openblas
- libcblas=3.8.0=17_openblas
- libclang=9.0.1=default_hde54327_0
- libffi=3.2.1=he1b5a44_1007
- libgcc-ng=9.2.0=h24d8f2e_2
- libgd=2.2.5=h307a58e_1007
- libgfortran-ng=7.5.0=hdf63c60_6
- libiconv=1.15=h516909a_1006
- liblapack=3.8.0=17_openblas
- libllvm9=9.0.1=he513fc3_1
- libopenblas=0.3.10=h5ec1e0e_0
- libpng=1.6.37=hed695b0_1
- libstdcxx-ng=9.2.0=hdf63c60_2
- libtiff=4.1.0=hc3755c2_3
- libuuid=2.32.1=h14c3975_1000
- libwebp=1.0.2=h56121f0_5
- libxcb=1.13=h14c3975_1002
- libxkbcommon=0.10.0=he1b5a44_0
- libxml2=2.9.10=hee79883_0
- llvm-openmp=10.0.0=hc9558a2_0
- lz4-c=1.9.2=he1b5a44_1
- matplotlib=3.2.2=0
- matplotlib-base=3.2.2=py36hb8e4980_0
- ncurses=6.1=hf484d3e_1002
- nspr=4.26=he1b5a44_0
- nss=3.47=he751ad9_0
- numpy=1.18.5=py36h7314795_0
- openjdk=11.0.1=h600c080_1018
- openssl=1.1.1g=h516909a_0
- pcre=8.44=he1b5a44_0
- perl=5.26.2=h516909a_1006
- perl-autoloader=5.74=pl526_2
- perl-carp=1.38=pl526_3
- perl-clone=0.42=pl526h516909a_0
- perl-config-general=2.63=pl526_0
- perl-digest-perl-md5=1.9=pl526_1
- perl-dynaloader=1.25=pl526_1
- perl-exporter=5.72=pl526_1
- perl-exporter-tiny=1.002001=pl526_0
- perl-extutils-makemaker=7.36=pl526_1
- perl-font-ttf=1.06=pl526_0
- perl-gd=2.71=pl526he860b03_0
- perl-io-string=1.08=pl526_3
- perl-list-moreutils=0.428=pl526_1
- perl-list-moreutils-xs=0.428=pl526_0
- perl-math-bezier=0.01=pl526_1
- perl-math-round=0.07=pl526_1
- perl-math-vecstat=0.08=pl526_1
- perl-module-implementation=0.09=pl526_2
- perl-module-runtime=0.016=pl526_1
- perl-number-format=1.75=pl526_3
- perl-params-validate=1.29=pl526h14c3975_1
- perl-pathtools=3.75=pl526h14c3975_1
- perl-readonly=2.05=pl526_0
- perl-regexp-common=2017060201=pl526_0
- perl-scalar-list-utils=1.52=pl526h516909a_0
- perl-set-intspan=1.19=pl526_1
- perl-statistics-basic=1.6611=pl526_2
- perl-svg=2.84=pl526_0
- perl-text-format=0.59=pl526_2
- perl-time-hires=1.9760=pl526h14c3975_1
- perl-try-tiny=0.30=pl526_1
- perl-xml-parser=2.44_01=pl526ha1d75be_1002
- perl-xsloader=0.24=pl526_0
- pip=20.1.1=py_1
- pthread-stubs=0.4=h14c3975_1001
- pyparsing=2.4.7=pyh9f0ad1d_0
- pyqt=5.12.3=py36haa643ae_3
- python=3.6.10=h8356626_1011_cpython
- python-dateutil=2.8.1=py_0
- python_abi=3.6=1_cp36m
- qt=5.12.5=hd8c4c69_1
- quast=5.0.2=py36pl526ha92aebf_0
- readline=8.0=hf8c457e_0
- setuptools=47.3.1=py36h9f0ad1d_0
- simplejson=3.17.0=py36h8c4c3a4_1
- six=1.15.0=pyh9f0ad1d_0
- sqlite=3.32.3=hcee41ef_0
- tk=8.6.10=hed695b0_0
- tornado=6.0.4=py36h8c4c3a4_1
- wheel=0.34.2=py_1
- xorg-fixesproto=5.0=h14c3975_1002
- xorg-inputproto=2.3.2=h14c3975_1002
- xorg-kbproto=1.0.7=h14c3975_1002
- xorg-libx11=1.6.9=h516909a_0
- xorg-libxau=1.0.9=h14c3975_0
- xorg-libxdmcp=1.1.3=h516909a_0
- xorg-libxext=1.3.4=h516909a_0
- xorg-libxfixes=5.0.3=h516909a_1004
- xorg-libxi=1.7.10=h516909a_0
- xorg-libxrender=0.9.10=h516909a_1002
- xorg-libxtst=1.2.3=h516909a_1002
- xorg-recordproto=1.14.2=h516909a_1002
- xorg-renderproto=0.11.1=h14c3975_1002
- xorg-xextproto=7.3.0=h14c3975_1002
- xorg-xproto=7.0.31=h14c3975_1007
- xz=5.2.5=h516909a_0
- zlib=1.2.11=h516909a_1006
- zstd=1.4.4=h6597ccf_3
- pip:
- pyqt5-sip==4.19.18
- pyqtchart==5.12
- pyqtwebengine==5.12.1
workflow/rules/analysis.smk
+ 56
− 56
View file @ 7bc8732e
......@@ -105,61 +105,61 @@ rule analysis_cdhit:
##################################################
# MMseqs2
rule analysis_mmseqs2_db:
input:
os.path.join(RESULTS_DIR, "assembly/{rtype}/{tool}/ASSEMBLY.fasta")
output:
os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype}/{tool}/ASSEMBLY_db")
log:
out="logs/mmseqs2_db.{rtype}.{tool}.out.log",
err="logs/mmseqs2_db.{rtype}.{tool}.err.log"
wildcard_constraints:
rtype="|".join(READ_TYPES),
tool="|".join(ASSEMBLERS)
conda:
os.path.join(ENV_DIR, "mmseqs2.yaml")
message:
"Create MMseqs2 DB from {input}"
shell:
"(date && mmseqs createdb {input} {output} && date) 2> {log.err} > {log.out}"
# rule analysis_mmseqs2_db:
# input:
# os.path.join(RESULTS_DIR, "assembly/{rtype}/{tool}/ASSEMBLY.fasta")
# output:
# os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype}/{tool}/ASSEMBLY_db")
# log:
# out="logs/mmseqs2_db.{rtype}.{tool}.out.log",
# err="logs/mmseqs2_db.{rtype}.{tool}.err.log"
# wildcard_constraints:
# rtype="|".join(READ_TYPES),
# tool="|".join(ASSEMBLERS)
# conda:
# os.path.join(ENV_DIR, "mmseqs2.yaml")
# message:
# "Create MMseqs2 DB from {input}"
# shell:
# "(date && mmseqs createdb {input} {output} && date) 2> {log.err} > {log.out}"
rule analysis_mmseqs2_compare:
input:
db1=os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype1}/{tool1}/ASSEMBLY_db"),
db2=os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype2}/{tool2}/ASSEMBLY_db")
output:
os.path.join(RESULTS_DIR, "analysis/mmseqs2/comparison/{rtype1}_{tool1}__{rtype2}_{tool2}")
log:
out="logs/mmseqs2.{rtype1}.{tool1}.{rtype2}.{tool2}.out.log",
err="logs/mmseqs2.{rtype1}.{tool1}.{rtype2}.{tool2}.err.log"
wildcard_constraints:
rtype1="|".join(READ_TYPES),
rtype2="|".join(READ_TYPES),
tool1="|".join(ASSEMBLERS),
tool2="|".join(ASSEMBLERS)
threads:
config["mmseqs2"]["threads"]
conda:
os.path.join(ENV_DIR, "cd-hit.yaml")
message:
"Create MMseqs2 compare: {input}"
shell:
# TODO: "mmseqs2_tmp" ??? (see old files)
"(date && mmseqs rbh {input.db1} {input.db2} {output} --min-seq-id 0.9 --threads {threads} && date) 2> {log.err} > {log.out}"
# rule analysis_mmseqs2_compare:
# input:
# db1=os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype1}/{tool1}/ASSEMBLY_db"),
# db2=os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype2}/{tool2}/ASSEMBLY_db")
# output:
# os.path.join(RESULTS_DIR, "analysis/mmseqs2/comparison/{rtype1}_{tool1}__{rtype2}_{tool2}")
# log:
# out="logs/mmseqs2.{rtype1}.{tool1}.{rtype2}.{tool2}.out.log",
# err="logs/mmseqs2.{rtype1}.{tool1}.{rtype2}.{tool2}.err.log"
# wildcard_constraints:
# rtype1="|".join(READ_TYPES),
# rtype2="|".join(READ_TYPES),
# tool1="|".join(ASSEMBLERS),
# tool2="|".join(ASSEMBLERS)
# threads:
# config["mmseqs2"]["threads"]
# conda:
# os.path.join(ENV_DIR, "cd-hit.yaml")
# message:
# "Create MMseqs2 compare: {input}"
# shell:
# # TODO: "mmseqs2_tmp" ??? (see old files)
# "(date && mmseqs rbh {input.db1} {input.db2} {output} --min-seq-id 0.9 --threads {threads} && date) 2> {log.err} > {log.out}"
rule analysis_mmseqs2_m8_convert:
input:
db1=os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype1}/{tool1}/ASSEMBLY_db"),
db2=os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype2}/{tool2}/ASSEMBLY_db"),
rbh=os.path.join(RESULTS_DIR, "analysis/mmseqs2/comparison/{rtype1}_{tool1}__{rtype2}_{tool2}")
output:
os.path.join(RESULTS_DIR, "analysis/mmseqs2/comparison/{rtype1}_{tool1}__{rtype2}_{tool2}.m8")
log:
out="logs/mmseqs2_convert.{rtype1}.{tool1}.{rtype2}.{tool2}.out.log",
err="logs/mmseqs2_convert.{rtype1}.{tool1}.{rtype2}.{tool2}.err.log"
conda:
os.path.join(ENV_DIR, "cd-hit.yaml")
message:
"Create MMseqs2 compare: {input}"
shell:
"(date && mmseqs convertalis {input.db1} {input.db2} {input.rbh} {output} && date) 2> {log.err} > {log.out}"
# rule analysis_mmseqs2_m8_convert:
# input:
# db1=os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype1}/{tool1}/ASSEMBLY_db"),
# db2=os.path.join(RESULTS_DIR, "analysis/mmseqs2/{rtype2}/{tool2}/ASSEMBLY_db"),
# rbh=os.path.join(RESULTS_DIR, "analysis/mmseqs2/comparison/{rtype1}_{tool1}__{rtype2}_{tool2}")
# output:
# os.path.join(RESULTS_DIR, "analysis/mmseqs2/comparison/{rtype1}_{tool1}__{rtype2}_{tool2}.m8")
# log:
# out="logs/mmseqs2_convert.{rtype1}.{tool1}.{rtype2}.{tool2}.out.log",
# err="logs/mmseqs2_convert.{rtype1}.{tool1}.{rtype2}.{tool2}.err.log"
# conda:
# os.path.join(ENV_DIR, "cd-hit.yaml")
# message:
# "Create MMseqs2 compare: {input}"
# shell:
# "(date && mmseqs convertalis {input.db1} {input.db2} {input.rbh} {output} && date) 2> {log.err} > {log.out}"
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment