Sequana

---------

* NEWS:

* added metropolis hastings module
* added a sniffer module for BAM/SAM/CRAM
* added a SMA/CRAM reader

* CHANGES:

* refactoring of bamtools. added SAM and CRAM classes. remove the
plot_acgt_content method. Instead of inheriting from pysam.Alignement,
we store the data as an attribute.

* FIXES:

* cutadapt rules and expdesign can now handle sample names with several
underscores
* Issue 515: sequanix should now be able to handle list in YAML files
* Issues 520: level info in sequanix was always set to INFO at start time
* Issue 519: fix issues in sequanix due to different API in new ruamel.yaml version
* Issue 522: fix bam_splitter tool

------

* BUGS:

* add /1 and /2 in quality control pipeline https://github.com/sequana/sequana/issues/508
* Fix test failure due to freebayes version 1 and 1.2 https://github.com/sequana/sequana/issues/512
* Fix reading of SampleSheet for MiSeq: https://github.com/sequana/sequana/issues/511
* Add Exp Design checked in quality control pipeline: https://github.com/sequana/sequana/issues/500

* CHANGES:

* sequana_vcf_filter: finalised version with INDEL removal, filters on DP4
and AF1 fields
* rename PacbioBAM into PacbioSubreads

0.6.X
-----

~~~~~~~~~~~

* CHANGES:

* sequana_coverage. Major refactoring of bedtools module to handle large
data sets (human), and provide ability to focus on CNVs using an
additional naive clustering (merge_rois_into_cnvs method) and binning.
We can also analyse data chunk by chunk (to avoid filling the memory).
added a plot_rois function
* sequana_coverage standalone: add the --chunksize, --cnv-clustering and
--binning options.

* NEWS:

* add cnvnator class
* coverage pipeline added in the pipelines

* BUGS:

* Fix silent warning (regex) in snpeff module
* double indexing adapters issue for Nextera fixed: https://github.com/sequana/sequana/issues/501

~~~~~~~~~~

* BUGS:

* Fix issue https://github.com/sequana/sequana/issues/380 is_sorted property
of the BAM class.
* Fix --no-report option in sequana_coverage and add --clustering (double
threshold option)
* pacbio_qc pipeline is now able to also read old pacbio format

* NEWS:

* SARTools rule added and used in the RNAseq pipeline
* add summary module to store summary in json formats.
* simple vcf_filter standalone

* CHANGES:

* pin kraken version to 1.1 (newest on bioconda)
* MAJOR REFACTORING of bedtools and sequana_coverage standalone. In
particular, change default window size to 20,001 or a fifth of genome
length (for small genome); speed up code; add plot_roi function, uses
multiqc for summary page; add log2 ratio column. See
https://github.com/sequana/sequana/issues/495 for details.
Scan large files by chunk. Add a snakemake that can be used in sequanix.
* remove the sequana_report standalone, which was not finalised and won't be
used in the future. We will use multiqc instead.

~~~~~~~~~~~~~

- a bug fix in the sequanix GUI and singularity. a statement related to the
ruamel.yaml package causes trouble if version is not 0.15. A temporary fix
consisted in adding a try/except (the statement is just a warning.filter and
has no impact on analysis)

~~~~~~~~~~~

* BUGS:

* Fix bug in the copy of the fastqc data sets in the quality control
* atropos bug in the reports (not full). Bug reported to atropos
github. https://github.com/jdidion/atropos/issues/57. Need to use version
1.1.16
* kraken report table were not sorted by percentage (as expected). Also,
if the case of poor databases with few entries, the output may contain
lots of classified sequences with Taxon 1, which was not reported
correctly in the krona plot.

* NEWS:

* example of a schema.yaml implemented for the quality control.
* sequanix: reads schema.yaml automatically for sequana pipelines
and can import one for generic cases. An option in the preference
was added to switch on/off the validation of the config file with this
schema. Can also import schema file for the generic case.

* CHANGES:

* Taxonomy file is downloaded for Kraken only when Kraken is used, not in
the main __init__ file anymore.