Re: [Seqan-dev] SeqAn Release 3.0.1

[Rahn, René <Rene.Rahn@fu-berlin.de>]

Dear Diaa Fayed,

thank you very much for writing us. We appreciate your feedback.
Can I ask you if there are features you are missing in particular? 

The goal of SeqAn3 is to reduce the code complexity with the help of modern C++, while still providing cutting edge algorithms and data structures for the bioinformatics community.
The implied changes to an existing program might be overwhelming at first but we believe there is a long-term benefit from that. 
Some of these benefits include that errors are easier to track, coding can be much faster through easier interfaces and less boiler plate code and our design goes well with new features from the standard library to get the most out of your tools. 
The other reason is that SeqAn2 has reached its end-of-life. There will likely be a last bug-fix release that improves compatibility with SeqAn3, but I am afraid that no further resources can be invested into keeping it up to date or fixing newly-found issues.

Regarding the advanced C++ concepts that are fairly new: We are going to provide more and more teaching materials as well as try to be around on conferences to teach those concepts and introduce features of the library. We will inform you via our channels of any of those opportunities. 

Of course, we have to admit that at the moment not all interfaces or features are complete and that it might be that some interfaces might change until 3.1 is fixed. 
On the other hand we'd like to hear from the community what their experience with the new library is so we can learn from them, improve the interfaces and provide a better high-quality and sustainable software for the community. 

So if you are critical of these changes, we would be particularly interested in your feedback if you do have the chance to try out SeqAn3!

Yours sincerely,

The SeqAn Team

> On 22. Jan 2020, at 19:10, Diaa Fayed <dr.diaafayed@gmail.com> wrote:
> 
> - I read about both SeqAn3 and SeqAn2.
> - Documentation and some algorithms have not implemented yet in SeqAn3
> - SeqAn3 is based on very advanced C++ concepts I have not known about  yet
> 
> so
> 
> I prefer to use SeqAn2.4
> 
> what is your opinion?
> ------------------------------------------------------------------------------------------------
> 
> On 1/22/20, Seiler, Enrico Frank <enrico.seiler@fu-berlin.de> wrote:
>> We are excited to present a new update of our SeqAn library. This release
>> has been in the making for roughly half a year now and we are proud to
>> present some great new features and also a lot of improvements with respect
>> to runtime and usability. You can find a comprehensive list of the changes
>> in our
>> changelog<http://docs.seqan.de/seqan/3.0.1/about_changelog.html#autotoc_md198>.
>> 
>>  *   Get to know SeqAn3 with our
>> tutorials<http://docs.seqan.de/seqan/3-master-user/usergroup1.html>.
>>  *   Visit our API
>> documentation<http://docs.seqan.de/seqan/3.0.1/index.html>.
>>  *   See the porting
>> guide<http://docs.seqan.de/seqan/3-master-user/howto_porting.html> for some
>> help on porting from SeqAn2.
>> 
>> Note that 3.1.0 will be the first API stable release and interfaces in this
>> release might still change.
>> [tada] Notable new features
>> 
>>  *
>> We added support for type erasing
>> semialphabets<http://docs.seqan.de/seqan/3.0.1/interfaceseqan3_1_1semialphabet.html>
>> that allows you to manage semialphabets with the same alphabet size in one
>> container. This can have a big effect on your compile-time, in case you
>> don't drink as much [coffee]  as we do.
>>  *
>> We added parallel support for the alignment algorithm. You can now configure
>> the number of threads you want to use for the alignment computation.
>>  *
>> One to command them all: Our argument parser now supports subcommands, such
>> as git pull. A
>> How-to<https://docs.seqan.de/seqan/3.0.1/subcommand_arg_parse.html> will
>> guide you through setting this up for your tool.
>>  *
>> The performance of the I/O was improved to allow faster file reading.
>> Further, we added support for reading and writing the CIGAR string through
>> alignment files.
>>  *
>> We added several new ranges and views. Most notably, the
>> seqan3::views::kmer_hash view, which transforms a sequence into a range of
>> k-mer hashes efficiently. Another view of great practice is the
>> seqan3::views::to, which can be used to convert a view into a container. We
>> also added aseqan3::dynamic_bitset which is a dynamic version of the
>> std::bitset.
>>  *
>> Memory consumption of the (bidirectional) FM-Index for text collections was
>> reduced by 10%.
>> 
>> [:trollface:] Notable API changes
>> As much as we'd like to reduce inconsistencies between releases, we are
>> sometimes forced to change an interface either to improve usability or to
>> follow changes made by the ISO C++ committee.
>> 
>>  *   All our concepts are named in the snake_case style (e.g.
>> seqan3::WritableAlphabet -> seqan3::writable_alphabet)!
>>  *   The directory seqan3/range/view has been renamed to
>> seqan3/range/views.
>>  *   The namespace seqan3::view has been renamed to seqan3::views.
>>  *   The CMake variable SEQAN3_VERSION_STRING defined by
>> find_package(SEQAN3) was renamed to SEQAN3_VERSION.
>> 
>> You can find a comprehensive list of the changes in our
>> changelog<http://docs.seqan.de/seqan/3.0.1/about_changelog.html#autotoc_md198>.
>> [bug] Notable bug fixes
>> 
>>  *   Copying and moving the seqan3::fm_index and seqan3::bi_fm_index now
>> work properly.
>>  *   The translation table for nucleotide to amino acid translation was
>> corrected.
>>  *   The amino acid score matrices were corrected.
>> 
>> [electric_plug] External dependencies
>> 
>>  *   We now support ranges-v3 versions >= 0.10.0 and < 0.11.0, increasing
>> the previous requirement of >= 0.5.0 and < 0.6.0.
>>  *   We now support cereal version 1.3.0, increasing the previous
>> requirement of 1.2.2
>> 
>> 
> 
> _______________________________________________
> seqan-dev mailing list
> seqan-dev@lists.fu-berlin.de
> https://lists.fu-berlin.de/listinfo/seqan-dev