Merge branch 'feature-a/update-TODOs' into 'dev'

Feature a/update TODOs

See merge request !257

algo/tests/bugs/bug20160121.should-get

 !LAUNCH: $VIDJIL_DIR/$EXEC -s '#####-#####' -c clones -r 1 -g ../../../germline/homo-sapiens.g:IGH -t 0 -e 1 bug20160121.fa
 
 $ Sequences should not be segmented since they only contain J.
-f1:UNSEG only J/3' -> 2
+1:UNSEG only J/3' -> 2

algo/tests/bugs/bug2249.should-get

 !LAUNCH: $VIDJIL_DIR/$EXEC -E 1 -d -g $VIDJIL_DIR/germline -c segment bug2249.fa
 
 $ Should not have a TRDD2 (length 9) with 8 deletions (first seq)
-fe0:7 TRDD2*01 1
+f0:7 TRDD2.01 1
 
 $ Should not have TRDD2 (length 9) with 12 deletions (second seq)
-fe0:6 TRDD2*01 6
+f0:6 TRDD2.01 6
 
 $ Should not have IGHD5-24 (length 20) with 29 deletions
-fe0:17 IGHD5-24*01 12
+f0:17 IGHD5-24.01 12
 
 $ Should not have VDDDJ
 f0:VDDDJ

algo/tests/should-get-tests/ambiguous_representative2.should-get

@@ -6,7 +6,7 @@ $ Three clones should be found
 # All clones should have a representative on 100%  of the length
 # Each regex should appear twice as it also appears in the similarity matrix.
 
-$ Representative length of clone 1
+$ Representative length of clone 1 (cf. #1541)
 f2:clone1.*100% of
 
 $ Representative length of clone 2

algo/tests/should-get-tests/chimera.should-get

 !LAUNCH: $VIDJIL_DIR/$EXEC $VIDJIL_DEFAULT_OPTIONS -A -uU -g $VIDJIL_DIR/germline $VIDJIL_DATA/chimera-trg.fa
 
-$ Do not segment on IGL by chance
-f1:IGL .* -> .* 0
+$ Do not segment on IG/TR by chance
+12:(IG|TR).* -> .* 0
 
 $ Report the sequence as AMBIGUOUS
 f1:UNSEG ambiguous -> 1
-
+# We are unable to report it as ambiguous yet as we are in the case where
+# we have +J+J +V+V. The maximum point won't be in the middle, thus the
+# V or J e-value will be high.

algo/tests/should-get-tests/segment_simul.should-get

@@ -17,5 +17,5 @@ $ First sequence, easy segmentation (no error, few deletions at the windows, sma
 # Note that a second D (D3-16*01) can be detected (6 common nucleotides, or even 7 with an overlap on the first D)
 # This is tested in segment_simul.should-vdj
 
-$ seq1_polyT is the same as seq1 but starts and ends with 10Ts, the recombination is in the middle. start position should not be 0 and end position should not be length-1
+$ seq1_polyT is the same as seq1 but starts and ends with 10Ts, the recombination is in the middle. start position should not be 0 and end position should not be length-1; See #2138
 f1:^>seq1_polyT \\+ VDJ 11 83 88 100 105 150 IGHV5-10-1.0[1-4] 2/CTTC/3 IGHD1-14.01 1/GTTA/5 IGHJ5.01

algo/tests/should-get-tests/segmentation-D-read-number.should-get

+# See #3003
 
 # We launch three times a sequence of interest (buggy-D.fa) with a various
 # number of distinct reads (mutations randomly inserted in the middle).

algo/tests/should-get-tests/stanford-fuse.should-get

 !REQUIRES: python $VIDJIL_DIR/tools/check_python_version.py
-!LAUNCH: $VIDJIL_DIR/$EXEC $VIDJIL_DEFAULT_OPTIONS -z 0 -w 60 -g $VIDJIL_DIR/germline/homo-sapiens.g:IGH $VIDJIL_DATA/Stanford_S22.fasta ; python $VIDJIL_DIR/tools/fuse.py out/Stanford_S22.vidjil out/Stanford_S22.vidjil -o out/fused.data ; cat out/fused.data | python $VIDJIL_DIR/tools/format_json.py -1
+!LAUNCH: $VIDJIL_DIR/$EXEC $VIDJIL_DEFAULT_OPTIONS -r 1 -z 0 -w 60 -g $VIDJIL_DIR/germline/homo-sapiens.g:IGH $VIDJIL_DATA/Stanford_S22.fasta ; python $VIDJIL_DIR/tools/fuse.py out/Stanford_S22.vidjil out/Stanford_S22.vidjil -o out/fused.data ; cat out/fused.data | python $VIDJIL_DIR/tools/format_json.py -1
 
 $ Points list
 1:"original_names": \[".*data//Stanford_S22.fasta", ".*data//Stanford_S22.fasta"\]
@@ -7,6 +7,5 @@ $ Points list
 $ Most abundant window, twice, fused 
 1:"id": "CCACCTATTACTGTACCCGGGAGGAACAATATAGCAGCTGGTACTTTGACTTCTGGGGCC".*"reads": \[8, 8\].*"top": 2
 
-# Fails since 49046ca6b97, no more 'others'
-$ Windows that are not in the top 50
-f1:"window": "others-1", .* "size": [ 12962, 12962 ]
+$ Windows that are not in the bottom of the distribution
+1: "distribution": \{[^}]+ "0.00001": \[9061, 9061\]