GitLab

Feature/fix segmenter

See merge request !3

Correct a typo: from ``this'' to ``self'' (``this'' is not defined in inner
function).

See #2092
Follows b461f5cd and ce33bb24

when adding a file to a generic sample set the sample set was not
selected by default due to the id being set incorrectly.

The name in get_info_dict was set to the sample type instead of the
sample name.

When adding a sample_set the page was then redirected to the wrong
sample_set due to the wrong id being passed to the controller

editing permissions or deleting a patient was broken due to the wrong id
being passed to the controller.

Follows 88c6e924.
The test is still failing.

Follows 88c6e924.

icon-pin is now used to fix the segmenter, we use icon-dot-3 to show that some
points could be hidden

just a little more info on the init_db form.

in the same vein as setting the admin password for users, the web2py
admin password is not easy to change once web2py itself has been
installed. Therefore defaulting it to the same as the vidjil admin
passwords seems like a good option

We're moving towards distributing vidjil so it's important to allow
users to choose their admin email and password.

This property tells whether the segmenter should be fixed or not.
When changing this property, this updates the corresponding icon.

See #1978 and !3.

With this commit we cannot fix the segmenter anymore
(the property checked of the input fix-segmenter is not defined).
See next commit.

See #1978 and !3

This is due to the Aho-Corasick automaton where we now have an index load for
the V gene and for the J gene. There is only one matching kmer from IGLJ but
that's enough to have an evalue < 1 because the index load for IGLJ is very
low and because there are only two reads in the dataset.

We could put a -e 1e-1 on the command line to make the test pass. Would this
be hacky or not?

This test passed because the 2 matched the 241 of the read length. But this is
not what we wanted to test. We expect just one sequence to be ambiguous.

This is necessary (for instance) with xxx germlines.

Kmer indexes and Aho-Corasick don't have the same features.
While it is possible to deal with several seeds in Aho-Corasick it is not with a simple lookup table.
Therefore we can consider different seeds only in some cases.

Launching init several times will allocate several times score_bins and nb_scores
which will cause memory leaks.

This is particularly the case in the WindowExtractor() which calls BinReadStorage
on germline->code but we can have different germlines with the same germline->code.

The default index will be set to Aho-Corasick when we'll
consider that it is ready. For now, option -q (quick) can launch AC automaton

Makes it ready to support a parameter to choose the index.

Allows to build any type of index instead of relying on one specific type.

This has been moved in another file for better code separation.

It will be useful to determine the type of index to build at runtime.

The affectation length is longer with Aho-Corasick than with k-mer based
indexes. So we use the length of the read instead.

The length of the affectations must be the seed span and not the seed weight
as the dashes are replaced by actual letters.

Unit test added.