3081896501

Build Type

github

Committed by GitHub

Commit Message

<a href="https://github.com/sequana/sequana/commit/<a class=hub.com/sequana/sequana/commit/c7110881ace2ba802fced730fbc47266f8025fba">c7110881a<a href="https://github.com/sequana/sequana/commit/c7110881ace2ba802fced730fbc47266f8025fba">">Merge </a><a class="double-link" href="https://github.com/sequana/sequana/commit/<a class="double-link" href="https://github.com/sequana/sequana/commit/18d3c37d0a4f2f90174ec52cad01c621eb3bb1d6">18d3c37d0</a>">18d3c37d0</a><a href="https://github.com/sequana/sequana/commit/c7110881ace2ba802fced730fbc47266f8025fba"> into 15dd4303c">15dd4303c</a>

Pull Request Pull Request #763: Dev

Run Details

36 of 36 new or added lines in 7 files covered. (100.0%)

12080 of 17241 relevant lines covered (70.07%)

0.7 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

0.0

/sequana/scripts/main/rnadiff.py

#  This file is part of Sequana software
#
#  Copyright (c) 2016-2020 - Sequana Development Team
#
#  Distributed under the terms of the 3-clause BSD license.
#  The full license is in the LICENSE file, distributed with this software.
#
#  website: https://github.com/sequana/sequana
#  documentation: http://sequana.readthedocs.io
#
##############################################################################
import os
import sys

import click
import colorlog

from sequana import GFF3
from sequana.rnadiff import RNADesign, RNADiffAnalysis
from sequana.utils import config

from .utils import CONTEXT_SETTINGS, common_logger

logger = colorlog.getLogger(__name__)


def rnadiff_auto_batch_column(ctx, args, incomplete):
    if "--design" in args:
        dfile = args[args.index("--design")]
        if os.path.exists(dfile):
            d = RNADesign(dfile)
        else:
            logger.critical("You must have a valid design.csv file following --design")
            sys.exit(1)
    else:
        try:
            d = RNADesign("design.csv")
            logger.warning('Using local design.csv to infer the batch column')
        except FileNotFoundError:
            logger.critical("No default design.csv found. Please use --design YOUR_DESIGN.csv ")
            sys.exit(1)


    batch = (x for x in d.df.columns if x not in {"label", "condition"})
    if len(batch) == 0:
        logger.warning("No batch effect included in your design file")
    else:
        return [c for c in batch if incomplete in c[0]]


def teardown(workdir):
    # common function to be used by subcommands to store called command
    from pathlib import Path
    from easydev import mkdirs

    workdir = Path(workdir)
    mkdirs(workdir / ".sequana")
    with open(Path(workdir) / ".sequana" / "info.txt", "w") as fout:
        from sequana import version

        fout.write(f"# sequana version: {version}\n")
        fout.write(" ".join(["sequana"] + sys.argv[1:]))


@click.command(context_settings=CONTEXT_SETTINGS)
@click.option(
    "--annotation",
    type=click.Path(),
    default=None,
    help="""The annotation GFF file used to perform the feature count""",
)
@click.option(
    "--output-directory",
    type=click.Path(),
    default="rnadiff",
    help="""Output directory where are saved the results. Use --force if it exists already""",
)
@click.option(
    "--force/--no-force",
    default=False,
    help="If output directory exists, use this option to erase previous results",
)
@click.option(
    "--features",
    type=click.Path(),
    default="all_features.out",
    help="""The Counts from feature counts. This should be the output of the
            sequana_rnaseq pipeline all_features.out """,
)
# FIXME I think it would be better to have a single file with multiple columns
# for alternative condition (specified using the "condition" option)
@click.option(
    "--design",
    type=click.Path(),
    default="design.csv",
    help="""It should have been generated by sequana_rnaseq. If
not, it must be a comma separated file with two columns. One for the label to be
found in the --features file and one column with the condition to which it
belong. Extra columns can be added to add batch effet. With 3 replicates and 2 conditions,
it should look like:

\b
label,condition
WT1,WT
WT2,WT
WT3,WT
file1,cond1
fileother,cond1
""",
)
@click.option(
    "--condition",
    type=str,
    default="condition",
    help="""The name of the column in design.csv to use as condition
for the differential analysis. Default is 'condition'""",
)
@click.option(
    "--feature-name",
    default="gene",
    help="The feature name compatible with your GFF (default is 'gene')",
)
@click.option(
    "--attribute-name",
    default="ID",
    help="""The attribute used as identifier. Compatible with your GFF (default is 'ID')""",
)
@click.option(
    "--reference",
    type=click.Path(),
    default=None,
    help="""The reference to test DGE against. If provided, conditions not
            involving the reference are ignored. Otherwise all combinations are
            tested""",
)
@click.option(
    "--comparisons",
    type=click.Path(),
    default=None,
    help="""By default, if a reference is provided, all conditions versus that
reference are tested. If no reference, the entire combinatory is performed
(Ncondition * (Ncondition-1) / 2. In both case all condtions found in the
design file are used. If a comparison file is provided, only conditions found in
it will be used. """,
)
@click.option(
    "--cooks-cutoff",
    type=click.Path(),
    default=None,
    help="""if none, let DESeq2 choose the cutoff""",
)
@click.option(
    "--independent-filtering/--no-independent-filtering",
    default=False,
    help="""Do not perform independent_filtering by default. low counts may not
have adjusted pvalues otherwise""",
)
@click.option(
    "--beta-prior/--no-beta-prior",
    default=False,
    help="Use beta prior or not. Default is no beta prior",
)
@click.option(
    "--batch",
    type=str,
    default=None,
    help="""set the column name (in your design) corresponding to the batch
effect to be included in the statistical model as batch ~ condition""",
    shell_complete=rnadiff_auto_batch_column,
)
@click.option(
    "--fit-type",
    default="parametric",
    help="DESeq2 type of fit. Default is 'parametric'",
)
@click.option(
    "--minimum-mean-reads-per-gene",
    default=0,
    help="""Keeps genes that have an average number of reads greater or equal this value. This is the average across all
replicates and conditions. Not recommended if you have lots of conditions. By default all genes are kept""",
)
@click.option(
    "--minimum-mean-reads-per-condition-per-gene",
    default=0,
    help="Keeps genes that have an average number of reads greater or equal to this value in all conditions. By default all genes are kept",
)
@click.option(
    "--keep-all-conditions/--no-keep-all-conditions",
    default=False,
    help="""Even though sub set of comparisons are provided, keep all conditions
in the analysis and report only the provided comparisons""",
)
@click.option(
    "--hover-name",
    default=None,
    help="""In volcano plot, we set the hover name to Name if present in the GFF,
otherwise to gene_id if present, then locus_tag, and finally ID and gene_name. One can specify
a hover name to be used with this option""",
)
@click.option(
    "--report-only",
    is_flag=True,
    help="""If analysis was done, you may want to redo the HTML report only using this option""",
)
@click.option("--xticks-fontsize", default=10, help="""Reduce fontsize of xticks""")
@common_logger
def rnadiff(**kwargs):
    """Perform RNA-seq differential analysis and reporting.

        This command performs the differential analysis of feature counts using DESeq2.
        A HTML report is created as well as a set of output files, including summary
        tables of the analysis.

        This command performs the differential analysis of gene expression based
        on the output of feature counts tool. The expected input is a tabulated file
        which is the aggregation of feature counts for each sample. This file is
        produced by the Sequana RNA-seq pipeline (https://github.com/sequana/rnaseq).
        It is named all_features.out and looks like:

            Geneid   Chr Start End Strand Length BAM1  BAM2  BAM3  BAM4
            ENSG0001 1       1  10      +     10 120    130   140  150
            ENSG0002 2       1  10      +     10 120    130     0    0


        To perform this analysis, you will also need the GFF file used during the RNA-seq
        analysis. You also need a design file that give the correspondance
        between the sample names found in the feature_count file above and the
        conditions of your RNA-seq analysis. The design looks like:

            label,condition
            BAM1,condition_A
            BAM2,condition_A
            BAM3,condition_B
            BAM4,condition_B

        Here is an example:

    \b
            sequana rnadiff --annotation Lepto.gff
                --design design.csv --features all_features.out
                 --feature-name gene --attribute-name ID

        The feature-name is the feature that was used in your counting.
        The attribute-name is the main attribute to use in the HTML reports.
        Note however, that all attributes found in your GFF file are repored
        in the HTML page

        Batch effet can be included by adding a column in the design.csv file. For
        example if called 'day', you can take this information into account using
        '--batch day'

        By default, when comparing conditions, all combination are computed. If
        you have N conditions, we compute the N(N-1)/2 comparisons. The
        reference is automatically chosen as the last one found in the design
        file. In this example:

            label,condition
            BAM1,A
            BAM2,A
            BAM3,B
            BAM4,B

        we compare A versus B. If you do not want that behaviour, use
        '--reference A'.

        In a more complex design,

            label,condition
            BAM1,A
            BAM2,A
            BAM3,B
            BAM4,B
            BAM5,C
            BAM6,C

        The comparisons are A vs B, A vs C and B vs C.
        If you wish to perform different comparisons or restrict the
        combination, you can use a comparison input file. For instance, to
        perform the C vs A  and C vs B comparisons only, create this
        file (e.g. comparison.csv):

            alternative,reference
            C,A
            C,B

        and use '--comparison comparison.csv'.


    """
    from sequana import logger
    import pandas as pd
    from sequana.modules_report.rnadiff import RNAdiffModule

    logger.setLevel(kwargs["logger"])

    from easydev import cmd_exists

    if not cmd_exists("Rscript"):
        logger.critical(
            """Rscript not found; You will need R and the DESeq2 package to be installed. 
You may install it yourself or use damona using the rtools:1.0.0 image """
        )
        sys.exit(1)

    outdir = kwargs["output_directory"]
    feature = kwargs["feature_name"]
    attribute = kwargs["attribute_name"]

    if os.path.exists(outdir) and not kwargs["force"]:
        logger.error(f"{outdir} exist already. Use --force to overwrite")
        sys.exit(1)

    if kwargs["annotation"]:
        gff_filename = kwargs["annotation"]
        logger.info("Checking annotation file (feature and attribute)")
        gff = GFF3(gff_filename)
        for feat in feature.split(","):
            if feat not in gff.features:
                logger.error(f"{feature} not found in the GFF. Most probably a wrong feature name")
                sys.exit(1)
            attributes = gff.get_attributes(feat)
            if attribute not in attributes:
                logger.error(
                    f"{attribute} not found in the GFF for the provided feature ({feat}). Most probably a wrong feature name."
                    " Please change --attribute-name option or do not provide any GFF"
                )
                sys.exit(1)
    else:
        gff = None

    comparisons = kwargs["comparisons"]
    if comparisons:
        # use \s*,\s* to strip spaces
        compa_df = pd.read_csv(comparisons, sep="\s*,\s*", engine="python")
        comparisons = list(zip(compa_df["alternative"], compa_df["reference"]))

    logger.info(f"Differential analysis to be saved into ./{outdir}")
    for k in sorted(
        [
            "independent_filtering",
            "beta_prior",
            "batch",
            "cooks_cutoff",
            "fit_type",
            "reference",
        ]
    ):
        logger.info(f"  Parameter {k} set to : {kwargs[k]}")

    # The analysis is here
    r = RNADiffAnalysis(
        kwargs["features"],
        kwargs["design"],
        kwargs["condition"],
        keep_all_conditions=kwargs["keep_all_conditions"],
        batch=kwargs["batch"],
        comparisons=comparisons,
        reference=kwargs["reference"],
        fc_feature=feature,
        fc_attribute=attribute,
        outdir=outdir,
        gff=gff,
        cooks_cutoff=kwargs.get("cooks_cutoff"),
        independent_filtering=kwargs.get("independent_filtering"),
        beta_prior=kwargs.get("beta_prior"),
        fit_type=kwargs.get("fit_type"),
        minimum_mean_reads_per_gene=kwargs.get("minimum_mean_reads_per_gene"),
        minimum_mean_reads_per_condition_per_gene=kwargs.get("minimum_mean_reads_per_condition_per_gene"),
    )

    if not kwargs["report_only"]:
        try:
            logger.info(f"Running DGE. Saving results into {outdir}")
            results = r.run()
            results.to_csv(f"{outdir}/rnadiff.csv")
        except Exception as err:
            logger.error(err)
            logger.error(f"please see {outdir}/code/rnadiff.err file for errors")
            sys.exit(1)

    logger.info("Reporting. Saving in summary.html")
    # this define the output directory where summary.html is saved
    config.output_dir = outdir

    import seaborn as sns

    RNAdiffModule(
        outdir,
        gff=gff,
        fc_attribute=attribute,
        fc_feature=feature,
        alpha=0.05,
        log2_fc=0,
        condition=kwargs["condition"],
        annot_cols=None,
        pattern="*vs*_degs_DESeq2.csv",
        palette=sns.color_palette(desat=0.6, n_colors=13),
        hover_name=kwargs["hover_name"],
        pca_fontsize=6,
        xticks_fontsize=kwargs.get("xticks_fontsize", 10),
    )

    #
    # save info.txt with sequana version
    teardown(outdir)

1	# This file is part of Sequana software
2	#
3	# Copyright (c) 2016-2020 - Sequana Development Team
4	#
5	# Distributed under the terms of the 3-clause BSD license.
6	# The full license is in the LICENSE file, distributed with this software.
7	#
8	# website: https://github.com/sequana/sequana
9	# documentation: http://sequana.readthedocs.io
10	#
11	##############################################################################
12	import os	×
13	import sys	×
14
15	import click	×
16	import colorlog	×
17
18	from sequana import GFF3	×
19	from sequana.rnadiff import RNADesign, RNADiffAnalysis	×
20	from sequana.utils import config	×
21
22	from .utils import CONTEXT_SETTINGS, common_logger	×
23
24	logger = colorlog.getLogger(__name__)	×
25
26
27	def rnadiff_auto_batch_column(ctx, args, incomplete):	×
28	if "--design" in args:	×
29	dfile = args[args.index("--design")]	×
30	if os.path.exists(dfile):	×
31	d = RNADesign(dfile)	×
32	else:
33	logger.critical("You must have a valid design.csv file following --design")	×
34	sys.exit(1)	×
35	else:
36	try:	×
37	d = RNADesign("design.csv")	×
38	logger.warning('Using local design.csv to infer the batch column')	×
39	except FileNotFoundError:	×
40	logger.critical("No default design.csv found. Please use --design YOUR_DESIGN.csv ")	×
41	sys.exit(1)	×
42
43
44	batch = (x for x in d.df.columns if x not in {"label", "condition"})	×
45	if len(batch) == 0:	×
46	logger.warning("No batch effect included in your design file")	×
47	else:
48	return [c for c in batch if incomplete in c[0]]	×
49
50
51	def teardown(workdir):	×
52	# common function to be used by subcommands to store called command
53	from pathlib import Path	×
54	from easydev import mkdirs	×
55
56	workdir = Path(workdir)	×
57	mkdirs(workdir / ".sequana")	×
58	with open(Path(workdir) / ".sequana" / "info.txt", "w") as fout:	×
59	from sequana import version	×
60
61	fout.write(f"# sequana version: {version}\n")	×
62	fout.write(" ".join(["sequana"] + sys.argv[1:]))	×
63
64
65	@click.command(context_settings=CONTEXT_SETTINGS)	×
66	@click.option(	×
67	"--annotation",
68	type=click.Path(),
69	default=None,
70	help="""The annotation GFF file used to perform the feature count""",
71	)
72	@click.option(	×
73	"--output-directory",
74	type=click.Path(),
75	default="rnadiff",
76	help="""Output directory where are saved the results. Use --force if it exists already""",
77	)
78	@click.option(	×
79	"--force/--no-force",
80	default=False,
81	help="If output directory exists, use this option to erase previous results",
82	)
83	@click.option(	×
84	"--features",
85	type=click.Path(),
86	default="all_features.out",
87	help="""The Counts from feature counts. This should be the output of the
88	sequana_rnaseq pipeline all_features.out """,
89	)
90	# FIXME I think it would be better to have a single file with multiple columns
91	# for alternative condition (specified using the "condition" option)
92	@click.option(	×
93	"--design",
94	type=click.Path(),
95	default="design.csv",
96	help="""It should have been generated by sequana_rnaseq. If
97	not, it must be a comma separated file with two columns. One for the label to be
98	found in the --features file and one column with the condition to which it
99	belong. Extra columns can be added to add batch effet. With 3 replicates and 2 conditions,
100	it should look like:
101
102	\b
103	label,condition
104	WT1,WT
105	WT2,WT
106	WT3,WT
107	file1,cond1
108	fileother,cond1
109	""",
110	)
111	@click.option(	×
112	"--condition",
113	type=str,
114	default="condition",
115	help="""The name of the column in design.csv to use as condition
116	for the differential analysis. Default is 'condition'""",
117	)
118	@click.option(	×
119	"--feature-name",
120	default="gene",
121	help="The feature name compatible with your GFF (default is 'gene')",
122	)
123	@click.option(	×
124	"--attribute-name",
125	default="ID",
126	help="""The attribute used as identifier. Compatible with your GFF (default is 'ID')""",
127	)
128	@click.option(	×
129	"--reference",
130	type=click.Path(),
131	default=None,
132	help="""The reference to test DGE against. If provided, conditions not
133	involving the reference are ignored. Otherwise all combinations are
134	tested""",
135	)
136	@click.option(	×
137	"--comparisons",
138	type=click.Path(),
139	default=None,
140	help="""By default, if a reference is provided, all conditions versus that
141	reference are tested. If no reference, the entire combinatory is performed
142	(Ncondition * (Ncondition-1) / 2. In both case all condtions found in the
143	design file are used. If a comparison file is provided, only conditions found in
144	it will be used. """,
145	)
146	@click.option(	×
147	"--cooks-cutoff",
148	type=click.Path(),
149	default=None,
150	help="""if none, let DESeq2 choose the cutoff""",
151	)
152	@click.option(	×
153	"--independent-filtering/--no-independent-filtering",
154	default=False,
155	help="""Do not perform independent_filtering by default. low counts may not
156	have adjusted pvalues otherwise""",
157	)
158	@click.option(	×
159	"--beta-prior/--no-beta-prior",
160	default=False,
161	help="Use beta prior or not. Default is no beta prior",
162	)
163	@click.option(	×
164	"--batch",
165	type=str,
166	default=None,
167	help="""set the column name (in your design) corresponding to the batch
168	effect to be included in the statistical model as batch ~ condition""",
169	shell_complete=rnadiff_auto_batch_column,
170	)
171	@click.option(	×
172	"--fit-type",
173	default="parametric",
174	help="DESeq2 type of fit. Default is 'parametric'",
175	)
176	@click.option(	×
177	"--minimum-mean-reads-per-gene",
178	default=0,
179	help="""Keeps genes that have an average number of reads greater or equal this value. This is the average across all
180	replicates and conditions. Not recommended if you have lots of conditions. By default all genes are kept""",
181	)
182	@click.option(	×
183	"--minimum-mean-reads-per-condition-per-gene",
184	default=0,
185	help="Keeps genes that have an average number of reads greater or equal to this value in all conditions. By default all genes are kept",
186	)
187	@click.option(	×
188	"--keep-all-conditions/--no-keep-all-conditions",
189	default=False,
190	help="""Even though sub set of comparisons are provided, keep all conditions
191	in the analysis and report only the provided comparisons""",
192	)
193	@click.option(	×
194	"--hover-name",
195	default=None,
196	help="""In volcano plot, we set the hover name to Name if present in the GFF,
197	otherwise to gene_id if present, then locus_tag, and finally ID and gene_name. One can specify
198	a hover name to be used with this option""",
199	)
200	@click.option(	×
201	"--report-only",
202	is_flag=True,
203	help="""If analysis was done, you may want to redo the HTML report only using this option""",
204	)
205	@click.option("--xticks-fontsize", default=10, help="""Reduce fontsize of xticks""")	×
206	@common_logger	×
207	def rnadiff(**kwargs):
208	"""Perform RNA-seq differential analysis and reporting.
209
210	This command performs the differential analysis of feature counts using DESeq2.
211	A HTML report is created as well as a set of output files, including summary
212	tables of the analysis.
213
214	This command performs the differential analysis of gene expression based
215	on the output of feature counts tool. The expected input is a tabulated file
216	which is the aggregation of feature counts for each sample. This file is
217	produced by the Sequana RNA-seq pipeline (https://github.com/sequana/rnaseq).
218	It is named all_features.out and looks like:
219
220	Geneid Chr Start End Strand Length BAM1 BAM2 BAM3 BAM4
221	ENSG0001 1 1 10 + 10 120 130 140 150
222	ENSG0002 2 1 10 + 10 120 130 0 0
223
224
225	To perform this analysis, you will also need the GFF file used during the RNA-seq
226	analysis. You also need a design file that give the correspondance
227	between the sample names found in the feature_count file above and the
228	conditions of your RNA-seq analysis. The design looks like:
229
230	label,condition
231	BAM1,condition_A
232	BAM2,condition_A
233	BAM3,condition_B
234	BAM4,condition_B
235
236	Here is an example:
237
238	\b
239	sequana rnadiff --annotation Lepto.gff
240	--design design.csv --features all_features.out
241	--feature-name gene --attribute-name ID
242
243	The feature-name is the feature that was used in your counting.
244	The attribute-name is the main attribute to use in the HTML reports.
245	Note however, that all attributes found in your GFF file are repored
246	in the HTML page
247
248	Batch effet can be included by adding a column in the design.csv file. For
249	example if called 'day', you can take this information into account using
250	'--batch day'
251
252	By default, when comparing conditions, all combination are computed. If
253	you have N conditions, we compute the N(N-1)/2 comparisons. The
254	reference is automatically chosen as the last one found in the design
255	file. In this example:
256
257	label,condition
258	BAM1,A
259	BAM2,A
260	BAM3,B
261	BAM4,B
262
263	we compare A versus B. If you do not want that behaviour, use
264	'--reference A'.
265
266	In a more complex design,
267
268	label,condition
269	BAM1,A
270	BAM2,A
271	BAM3,B
272	BAM4,B
273	BAM5,C
274	BAM6,C
275
276	The comparisons are A vs B, A vs C and B vs C.
277	If you wish to perform different comparisons or restrict the
278	combination, you can use a comparison input file. For instance, to
279	perform the C vs A and C vs B comparisons only, create this
280	file (e.g. comparison.csv):
281
282	alternative,reference
283	C,A
284	C,B
285
286	and use '--comparison comparison.csv'.
287
288
289	"""
290	from sequana import logger	×
291	import pandas as pd	×
292	from sequana.modules_report.rnadiff import RNAdiffModule	×
293
294	logger.setLevel(kwargs["logger"])	×
295
296	from easydev import cmd_exists	×
297
298	if not cmd_exists("Rscript"):	×
299	logger.critical(	×
300	"""Rscript not found; You will need R and the DESeq2 package to be installed.
301	You may install it yourself or use damona using the rtools:1.0.0 image """
302	)
303	sys.exit(1)	×
304
305	outdir = kwargs["output_directory"]	×
306	feature = kwargs["feature_name"]	×
307	attribute = kwargs["attribute_name"]	×
308
309	if os.path.exists(outdir) and not kwargs["force"]:	×
310	logger.error(f"{outdir} exist already. Use --force to overwrite")	×
311	sys.exit(1)	×
312
313	if kwargs["annotation"]:	×
314	gff_filename = kwargs["annotation"]	×
315	logger.info("Checking annotation file (feature and attribute)")	×
316	gff = GFF3(gff_filename)	×
317	for feat in feature.split(","):	×
318	if feat not in gff.features:	×
319	logger.error(f"{feature} not found in the GFF. Most probably a wrong feature name")	×
320	sys.exit(1)	×
321	attributes = gff.get_attributes(feat)	×
322	if attribute not in attributes:	×
323	logger.error(	×
324	f"{attribute} not found in the GFF for the provided feature ({feat}). Most probably a wrong feature name."
325	" Please change --attribute-name option or do not provide any GFF"
326	)
327	sys.exit(1)	×
328	else:
329	gff = None	×
330
331	comparisons = kwargs["comparisons"]	×
332	if comparisons:	×
333	# use \s,\s to strip spaces
334	compa_df = pd.read_csv(comparisons, sep="\s,\s", engine="python")	×
335	comparisons = list(zip(compa_df["alternative"], compa_df["reference"]))	×
336
337	logger.info(f"Differential analysis to be saved into ./{outdir}")	×
338	for k in sorted(	×
339	[
340	"independent_filtering",
341	"beta_prior",
342	"batch",
343	"cooks_cutoff",
344	"fit_type",
345	"reference",
346	]
347	):
348	logger.info(f" Parameter {k} set to : {kwargs[k]}")	×
349
350	# The analysis is here
351	r = RNADiffAnalysis(	×
352	kwargs["features"],
353	kwargs["design"],
354	kwargs["condition"],
355	keep_all_conditions=kwargs["keep_all_conditions"],
356	batch=kwargs["batch"],
357	comparisons=comparisons,
358	reference=kwargs["reference"],
359	fc_feature=feature,
360	fc_attribute=attribute,
361	outdir=outdir,
362	gff=gff,
363	cooks_cutoff=kwargs.get("cooks_cutoff"),
364	independent_filtering=kwargs.get("independent_filtering"),
365	beta_prior=kwargs.get("beta_prior"),
366	fit_type=kwargs.get("fit_type"),
367	minimum_mean_reads_per_gene=kwargs.get("minimum_mean_reads_per_gene"),
368	minimum_mean_reads_per_condition_per_gene=kwargs.get("minimum_mean_reads_per_condition_per_gene"),
369	)
370
371	if not kwargs["report_only"]:	×
372	try:	×
373	logger.info(f"Running DGE. Saving results into {outdir}")	×
374	results = r.run()	×
375	results.to_csv(f"{outdir}/rnadiff.csv")	×
376	except Exception as err:	×
377	logger.error(err)	×
378	logger.error(f"please see {outdir}/code/rnadiff.err file for errors")	×
379	sys.exit(1)	×
380
381	logger.info("Reporting. Saving in summary.html")	×
382	# this define the output directory where summary.html is saved
383	config.output_dir = outdir	×
384
385	import seaborn as sns	×
386
387	RNAdiffModule(	×
388	outdir,
389	gff=gff,
390	fc_attribute=attribute,
391	fc_feature=feature,
392	alpha=0.05,
393	log2_fc=0,
394	condition=kwargs["condition"],
395	annot_cols=None,
396	pattern="vs_degs_DESeq2.csv",
397	palette=sns.color_palette(desat=0.6, n_colors=13),
398	hover_name=kwargs["hover_name"],
399	pca_fontsize=6,
400	xticks_fontsize=kwargs.get("xticks_fontsize", 10),
401	)
402
403	#
404	# save info.txt with sequana version
405	teardown(outdir)	×

sequana / sequana / 3081896501

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous