RELEASE_NOTES.txt

Release notes for FastQ Screen v0.16.0 (20 September 2024)
----------------------------------------------------------
Added functionality to process long sequence reads by using 
Minimap2 as the aligner.

We would like to thank Martin Pollard (Wellcome Sanger 
Institute, UK) for contributing this improvement.



Release notes for FastQ Screen v0.15.3 (03 April 2023)
------------------------------------------------------
Fixed filtering bug causing the final filtered FASTQ 
file to have the file extension .fastq when it should be 
.fastq.gz, since the file is actually gzipped.



Release notes for FastQ Screen v0.15.2 (26 January 2022)
--------------------------------------------------------
Updated documentation

Provided details on the genomes downloaded with the 
command --get_genomes for possible future reference.


Release notes for FastQ Screen v0.15.1 (11 January 2022)
--------------------------------------------------------
Updated contact details. 



Release notes for FastQ Screen v0.14.1 (03 June 2020)
----------------------------------------------------------
Added --score_min L,0,-0.6 as a Bismark/Bowtie2 mapping 
parameter to make FastQ Screen perform less stringent 
mapping, which is more consistent with non-bisulfite
FastQ Screen mapping.

Added a GitHUb Actions YAML file for testing FastQ Screen.



Release notes for FastQ Screen v0.14.0 (17 June 2019)
----------------------------------------------------------
New --add_genome option.

Fixed bug causing FastQ Screen to run when aligner not present.



Release notes for FastQ Screen v0.13.0 (12 September 2018)
----------------------------------------------------------
Added --get_genomes option.

Updated documentation.



Release notes for FastQ Screen v0.12.2 (16 August 2018)
-------------------------------------------------------
For bisulfite mapping, bismark is now run in --ambiguous 
mode to identify multi-mapping reads.



Release notes for FastQ Screen v0.12.1 (24 July 2018)
-----------------------------------------------------
Added --inverse option for filtering files.

Improved results display format.



Release notes for FastQ Screen v0.12.0 (15 June 2018)
-----------------------------------------------------
Interactive HTML graphs now made using Plotly.



Release notes for FastQ Screen v0.11.4 (27 November 2017)
---------------------------------------------------------
FASTQ files are edited so that the third line of a read 
is always a plus symbol.  This prevents tagged/filtered 
output files not technically adhering to FASTQ format. 



Release notes for FastQ Screen v0.11.3 (16 October 2017)
--------------------------------------------------------
FastQ Screen uses full path to dependencies rather than Bowtie,
Bowtie2 etc.



Release notes for FastQ Screen v0.11.2 (21 September 2017)
----------------------------------------------------------
Fixed bug preventing --tag being selected without --filter.

In bisulfite mode, FastQ Screen no longer assumes that 
Bowtie/Bowtie2 are always in the path (even if specified otherwise 
in the config file).

FastQ Screen now terminates before creating a subset file if no 
aligner/Bismark executable file is found.

FastQ Screen no longer gives an initialisation warning if, in the 
configuration file, the DATABASE line does not specify a database 
name and/or database path.



Release notes for FastQ Screen v0.11.1 (23 February 2017)
---------------------------------------------------------
v0.11.1 is a minor release.

Fixed bug preventing user selecting --filter option 4 or 5.



Release notes for FastQ Screen v0.11.0 (22 February 2017)
---------------------------------------------------------
v0.11.0 is a major release.

Added --filter options 4 and 5.  

Added option --pass to further improve filtering.

Prevented HTML bar graphs overlapping when screening against
multiple genomes.

Corrected documentation describing how to use the option --top.



Release notes for FastQ Screen v0.10.0 (19 January 2017)
-----------------------------------------------------------
v0.10.0 is a major release.

Added --top functionality for when faster processing times is the main 
priority.

Improved presentation of HTML graphs.



Release notes for FastQ Screen v0.9.5 (07 December 2016)
--------------------------------------------------------
v0.9.5 is a minor release.

Fixed bug causing FastQ Screen, when running in --bisulfite mode, to 
mislabel species names in the HTML bisulfite read orientation graph 
on the occasion that one or more of the samples tested contained 
reads that mapped to none of the bisulfite converted reference genomes.



Release notes for FastQ Screen v0.9.4 (21 November 2016)
--------------------------------------------------------
v0.9.4 is a minor release.

New colour scheme which should be easily interpretable by colour blind
people. 

Fixed bug preventing, in some instances, genome names being displayed 
in the header line of filtered (using the --filter option) FASTQ 
files.



Release notes for FastQ Screen v0.9.3 (31 October 2016)
-------------------------------------------------------
v0.9.3 is a minor release.

Fixed bug stopping the command-line --threads option overriding 
the configuration file. 

When the --tag option was specified, FastQ Screen should have 
analysed all the reads in a file by default. However, a bug resulted
in a subset file being generated and analysed instead. This has been
fixed and now a reduced reads file will only be generated with the 
--tag option if explicitly requested using the --subset command.

Fixed bug causing FastQ Screen to not process reads containing 
a full-stop in the FASTQ read header.  

Changes to what is reported when using the --quiet option

Updated documentation.



Release notes for FastQ Screen v0.9.2 (12 October 2016)
-------------------------------------------------------
v0.9.2 is a minor release.

When --outdir option selected, FastQ Screen creates the output 
directory if it does not already exist.

FastQ Screen in bisulfite mode checks whether the bisulfite 
orientation graph already exists.

Adjusted how compressed files are read to improve compatibility 
with Mac systems.

Fixed bug causing the HTML report to be missing one genome when 
reporting conventional (i.e. not bisulfite) alignment results

Updated documentation.



Release notes for FastQ Screen v0.9.1 (05 October 2016)
-------------------------------------------------------
v0.9.1 is a minor release.

Fixed bug causing FastQ Screen to terminate prematurely when in 
bisulfite mode if no reads map to a bisulfite reference genome.



Release notes for FastQ Screen v0.9.0 (04 October 2016)
-------------------------------------------------------
v0.9.0 is a major release.

FastQ Screen, when run in Bisulfite mode, reports to which strand 
reads aligned (original top strand, complementary to original top 
strand, complementary to original bottom strand, or original bottom 
strand).



Release notes for FastQ Screen v0.8.0 (08 September 2016)
---------------------------------------------------------
v0.8.0 is a major release.

Program is now compatible with aligner BWA

FastQ Screen produces an HTML summary report

Program documentation has been substantially updated and is now in 
Markdown format



Release notes for FastQ Screen v0.7.0 (01 August 2016)
------------------------------------------------------
v0.7.0 is a major release.

Added --tag option to create output FASTQ files in which the
the genomes to which a read maps is appended to the first line
of the FASTQ read.

Added --filter option to extract reads from a tagged FASTQ file
which map, or do not map, to a specified combination of genomes.

Pre-existing option --nohits is now equivalent to the parameters
--tag --filter 000 (number of zeroes corresponds to the number
of genome being screened)



Release notes for FastQ Screen v0.6.4 (12 July 2016)
----------------------------------------------------
v0.6.4 is a minor release.

Program no longer terminates if a single Bismark reference genome 
index is incorrectly specified.

Fixed bug causing program to crash if --aligner bowtie2 and 
--bisulfite specified together.

FastQ Screen can now use Bowtie (in addition to Bowtie2) when 
performing Bisulfite mapping with Bismark.

Fixed bug in how FastQ Screen checks for dependencies 
(e.g. SamTools).



Release notes for FastQ Screen v0.6.3 (07 July 2016)
----------------------------------------------------
v0.6.3 is a minor release.

Fixed bug causing --subset 0 to crash.

Fixed bug in which the reported percentage reads mapping to no 
libraries was, in some instances, an underestimate of the correct 
value.



Release notes for FastQ Screen v0.6.2 (05 July 2016)
----------------------------------------------------
v0.6.2 is a minor release.

Updated help text.

Refactored code.



Release notes for FastQ Screen v0.6.1 (01 July 2016)
----------------------------------------------------
v0.6.1 is a minor release.

Fixed bug causing program to crash in some instances when --outdir
option selected.



Release notes for FastQ Screen v0.6.0 (27 June 2016)
----------------------------------------------------
v0.6.0 is a major release.

Compatible with Bismark, enabling bisulfite library QC.

Use the option --bisulfite when processing bisulfite libraries. The 
path to Bismark may be specified in the configuration file. 
Bisulfite libraries cannot be processed simultaneously with 
non-bisulfite libraries. Bismark can be downloaded from:
www.bioinformatics.babraham.ac.uk/projects/bismark

Option --colorspace is no longer supported.



Release notes for FastQ Screen v0.5.2 (07 September 2015)
---------------------------------------------------------
v0.5.2 is a minor release.

Fixed bug observed when --nohits option selected causing 
initialization warnings, in some instances.



Release notes for FastQ Screen v0.5.1 (14 July 2015)
----------------------------------------------------
v0.5.1 is a minor release. Program checks the  configuration to 
ensure a FASTQ file is not mapped against the same library more 
than once.



Release notes for FastQ Screen v0.5.0 (29 June 2015)
----------------------------------------------------
v0.5.0 is a major release.

Please note that users no longer need to specify whether a genome 
index is compatible with bowtie or bowtie2, since this is now 
determined automatically.

Option --subset 100000 is now the default.  Use --subset 0 to 
process an entire file (not recommended for most QC applications, 
since this generally takes much more time). 

Option --paired removed. Mapping forward or reverse reads 
independently should be adequate to ascertain whether there is 
contamination, and will also provide the user with additional 
information if the forward reads are more prone generally to 
contamination than the reverse reads (or vice versa). Also, 
sometimes the script will fail to detect contamination in 
--paired mode. For example, if the read pair did not 
constitute a contiguous region of DNA, or if the paired reads 
were separated by are large distance (such as RNA seq).

Bowtie2 is now the default aligner, replacing the orignal bowtie.

New option --force instructs FastQ Screen to overwrite extant output 
files.

The script now uses a more memory efficient internal data structure
for recording which reads map to what library. However, this means
that a maximum of 15 libraries may be specified with 32-bit Perl
or 31 libraries with 64-bit Perl.



Release notes for FastQ Screen v0.4.4
-------------------------------------
v0.4.4 is a minor release.

Fixed a bug in the code which checked for the presence of bowtie2
indices for large genomes.

Added a nicer font for the graph and improved some of the colours

Fixed the output file naming for .fq files and files with names
ending in .gz.



Release notes for FastQ Screen v0.4.3
-------------------------------------
v0.4.3 is a minor release. Fixed bug causing all reads to be 
written to the 'no hits' output file when using Bowtie2 as the 
aligner. Bowtie2 now runs with the parameters '--no-discordant' 
and '--no-mixed' when mapping paired-end reads.  The 'nohits' 
output file has the file extension '.fastq' and is compressed if 
the input files are compressed.



Release notes for FastQ Screen v0.4.2
-------------------------------------
v0.4.2 is a minor release. The script no longer defaults to 
Bowtie if "--aligner" is not specified.  Instead, the script checks 
the configuration file to determine if Bowtie/Bowtie2 paths and 
indices  have been specified. If both Bowtie and Bowtie2 indices 
have been specified, FastqScreen then defaults to the original 
Bowtie. The script now reports the number of reads mapping each 
genome in addition to percentages.



Release notes for FastQ Screen v0.4.1
-------------------------------------
v0.4.1 is a minor release. A new command line argument has been 
introduced (--aligner, -a) which enables the user to select the 
sequence aligner to use; valid options are 'bowtie' (default) or 
'bowtie2'. Fastq_screen now only uses one aligner per round of 
screening and so if the configuration file lists paths to both bowtie 
and bowtie2 libraries, only the libraries for the specified aligner 
will be imported.

Other changes:
* The script will skip the production of graphs if GD::Graph is not 
installed to prevent terminating in error.

* Standard error is now reported as a separate file for each library.



Release notes for FastQ Screen v0.4
-----------------------------------
v0.4 is a major release enabling FastQ Screen to use Bowtie2 as the 
sequence aligner. Bowtie2 may be preferred since it is more efficient 
than its predecessor at processing longer reads and can perform gapped 
alignments. The configuration file instructs FastQ Screen as to which 
aligner to use.  

We would like to thank Patrick J Biggs (Institute of Veterinary, 
Animal & Biomedical Sciences, Massey University, New Zealand) and 
Saulo Alves Aflitos (Plant Research International, Wageningen, The 
Netherlands) for suggesting this improvement.



Release notes for FastQ Screen v0.3.1
-------------------------------------
v0.3.1 is a minor release fixing a bug that prevented the script
receiving options using the Bowtie argument (--bowtie, -b).



Release notes for FastQ Screen v0.3
-----------------------------------
v0.3 has new data analysis, output and file handling functionality.

The most significant change is the removal of the multilib option.
The script now calculates the percentage of sequences that map:
i) uniquely to the specified genome but no others; ii) more than 
once to the specified genome but to no others; iii) once to the
specified genome and to other libraries; iv) more than once to
the specified genome and other libraries as well; iv) to none of
the reference genomes. Essentially, this combines the 
functionality of selecting/not selecting multilib in the previous 
version of FastQ Screen.

Other changes:
* A new option (--nohits) prints to an output file sequence reads or 
read pairs that mapped to none of the reference genomes.  If used in 
conjunction with the subset option, only reads/read pairs present in 
the temporary subset file that failed to map to any genome will be 
returned.

* The script can now process zipped FASTQ files with the file extension 
'gz'.

* The --illumina flag has changed to --illumina1_3 in recognition of
current Illumina sequencers now reporting in Sanger format.

* The script can now process files with names that contain spaces. 



Release notes for FastQ Screen v0.2.1
-------------------------------------
v0.2.1 is a minor release which fixes a bug in paired end multilib
searches where the mapped percentage values reported are double the
true values (leading to percentages >100% and corrupted graphs).

Single end searches and searches using the default mode (rather than
multilib) were unaffected.



Release notes for FastQ Screen v0.2
-----------------------------------
v0.2 adds some new features and a new mode of analysis.

The major change is a new way of analysing data called multilib mode.
This mode is activated by passing the --multilib argument to the 
program on the command line.  In this mode each individual sequence
searched is tracked across all search libraries.  The single and
multiple hits in the output therefore say how many sequences hit
this, and only this, library, and how many hit this library and one
or more other libraries.  In multilib mode it doesn't matter how
many times a sequence could have been mapped in a given library, only
whether it found a hit or not.

Because multilib mode needs to record the pattern of hits for every
single sequence searched it will have significantly higher memory
usage than the default search mode, especially if run without the
--subset option enabled.

Other changes in this release are:

* The addition of colorspace support via the --color option. To use
  this option your bowtie indices will need to have been generated
  with the colorspace flag set.

* A new option (--conf) to manually specify a location for the config
  file rather than using the default file installed with the program.
  Makes it easy to switch between different sets of libraries.

* A new option (--threads) to override the default setting in the conf
  file for how many threads to parallelise your searches across.