tests: should-get-tests, replace '../..' with '$VIDJIL_DIR'

This will be more flexible.

algo/tests/should-get-tests/10-md5-data.should-get

 !NO_LAUNCHER:
-!LAUNCH: (cd ../../data ; md5sum *.fasta || md5 -r *.fasta)
+!LAUNCH: (cd $VIDJIL_DIR/data ; md5sum *.fasta || md5 -r *.fasta)
 
 $ Check md5 in data/
 1:985219d9156b6d9e8966807595beada3 Stanford_S22.fa

algo/tests/should-get-tests/10-md5-germline.should-get

 !NO_LAUNCHER:
-!LAUNCH: (cd ../../germline ; md5sum *.fa || md5 -r *.fa)
+!LAUNCH: (cd $VIDJIL_DIR/germline ; md5sum *.fa || md5 -r *.fa)
 
 $ Check md5 in germline/, sequences split and processed from IMGT
 1:3a655c9d99ca04907a120f1b69febf2e IGHD.fa

algo/tests/should-get-tests/cgi-align.should-get

-!LAUNCH: export REQUEST_METHOD=POST ; export CONTENT_TYPE=test ; $LAUNCHER ../../browser/cgi/align.cgi < ../../data/msa.fa
+!LAUNCH: export REQUEST_METHOD=POST ; export CONTENT_TYPE=test ; $LAUNCHER $VIDJIL_DIR/browser/cgi/align.cgi < $VIDJIL_DIR/data/msa.fa
 
 $ no spurious info
 0: .* bp in .* sequences

algo/tests/should-get-tests/chimera-fake-2.should-get

-!LAUNCH: ../../vidjil -A -t 0 -g ../../germline -2 -i ../../data/chimera-fake.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -A -t 0 -g $VIDJIL_DIR/germline -2 -i $VIDJIL_DIR/data/chimera-fake.fa
 
 $ The KmerSegmenter segments the three chimera reads on PSEUDO_MAX12 germline (-2)
 1:xxx .* -> .* 3

algo/tests/should-get-tests/chimera-fake-VJ-germlines.should-get

-!LAUNCH: ../../vidjil -A -g ../../data/chimera-fake-VJ-germlines.data -i ../../data/chimera-fake-VJ.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -A -g $VIDJIL_DIR/data/chimera-fake-VJ-germlines.data -i $VIDJIL_DIR/data/chimera-fake-VJ.fa
 # Testing a custom (fake) germlines.data
 
 $ Loads from 'chimera-fake-VJ-germlines.data' a custom germline

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

-!LAUNCH: ../../vidjil -A -g ../../germline -i -2 ../../data/chimera-fake-VJ.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -A -g $VIDJIL_DIR/germline -i -2 $VIDJIL_DIR/data/chimera-fake-VJ.fa
 
 $ The KmerSegmenter segments the five chimera reads on PSEUDO_MAX12 germline (-2)
 1:xxx .* -> .* 5

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

-!LAUNCH: ../../vidjil -e 100 -A -t 0 -g ../../germline -4 -i ../../data/chimera-fake-half.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -e 100 -A -t 0 -g $VIDJIL_DIR/germline -4 -i $VIDJIL_DIR/data/chimera-fake-half.fa
 # TODO: a more precise modeling should give a e-value computation that could make this work even with -e 1
 
 $ The KmerSegmenter segments the six chimera reads on PSEUDO_MAX1U germline (-4)

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

-!LAUNCH: ../../vidjil -g ../../germline -i ../../data/chimera-fake.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -g $VIDJIL_DIR/germline -i $VIDJIL_DIR/data/chimera-fake.fa
 
 $ Do not segment on any germline, even incomplete
 1:junction detected in 0 reads

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

-!LAUNCH: ../../vidjil -A -uU -g ../../germline ../../data/chimera-trg.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -A -uU -g $VIDJIL_DIR/germline $VIDJIL_DIR/data/chimera-trg.fa
 
 $ Do not segment on IGL by chance
 1:IGL .* -> .* 0

algo/tests/should-get-tests/clones_lec.should-get

-!LAUNCH: ../../vidjil -G ../../germline/TRG -c clones -A -3 ../../data/segment_lec.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -G $VIDJIL_DIR/germline/TRG -c clones -A -3 $VIDJIL_DIR/data/segment_lec.fa
 
 $ Extract 50bp windows (TRG)
 1:found . 50-windows

algo/tests/should-get-tests/clones_simul.should-get

-!LAUNCH: ../../vidjil -k 14 -w 50 -c clones -G ../../germline/IGH -y 3 -z 1 -r 1 ../../data/clones_simul.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -k 14 -w 50 -c clones -G $VIDJIL_DIR/germline/IGH -y 3 -z 1 -r 1 $VIDJIL_DIR/data/clones_simul.fa
 
 $ Junction extractions
 1:found 25 50-windows in 66 reads

algo/tests/should-get-tests/clones_simul_cluster.should-get

-!LAUNCH: ../../vidjil -k 14 -w 50 -c clones -G ../../germline/IGH -y 3 -z 0 -r 1 -n 5 ../../data/clones_simul.fa ; cat out/clones_simul.vidjil
+!LAUNCH: $VIDJIL_DIR/vidjil -k 14 -w 50 -c clones -G $VIDJIL_DIR/germline/IGH -y 3 -z 0 -r 1 -n 5 $VIDJIL_DIR/data/clones_simul.fa ; cat out/clones_simul.vidjil
 
 $ Window extractions
 2:found 25 50-windows in 66 reads

algo/tests/should-get-tests/common-V-D.should-get

-!LAUNCH: ../../vidjil -KA -z 0 -s \\\\#\\\\#\\\\#\\\\#\\\\#-\\\\#\\\\#\\\\#\\\\#\\\\# -G ../../germline/IGH  ../../data/common-V-D.fa  ; cat out/common-V-D.affects
+!LAUNCH: $VIDJIL_DIR/vidjil -KA -z 0 -s \\\\#\\\\#\\\\#\\\\#\\\\#-\\\\#\\\\#\\\\#\\\\#\\\\# -G $VIDJIL_DIR/germline/IGH  $VIDJIL_DIR/data/common-V-D.fa  ; cat out/common-V-D.affects
 
 $ Segments the sequence
 1: SEG  .* -> .* 1 

algo/tests/should-get-tests/gzip.should-get

 !NO_LAUNCHER:
-!LAUNCH: ../../vidjil -G ../../germline/IGH ../../data/clones_simul.fa > out-fa ;  ../../vidjil -G ../../germline/IGH -b clones_simul ../../data/clones_simul.fa.gz > out-fa-gz ; diff -s -I '\#' -I 'index' -I 'data/clones_simul' out-fa out-fa-gz ; echo 'Diff: '\\$?; wc -l out-fa-gz
+!LAUNCH: $VIDJIL_DIR/vidjil -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/clones_simul.fa > out-fa ;  $VIDJIL_DIR/vidjil -G $VIDJIL_DIR/germline/IGH -b clones_simul $VIDJIL_DIR/data/clones_simul.fa.gz > out-fa-gz ; diff -s -I '\#' -I 'index' -I 'data/clones_simul' out-fa out-fa-gz ; echo 'Diff: '\\$?; wc -l out-fa-gz
 
 $ Identical output
 1:Diff: 0

algo/tests/should-get-tests/large-N.should-get

-!LAUNCH: ../../vidjil -G ../../germline/IGH -r 1 ../../data/large_N.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -G $VIDJIL_DIR/germline/IGH -r 1 $VIDJIL_DIR/data/large_N.fa
 
 $ Find a huge insertion in the segmentation
 1:>clone-001.*/[ACGT]{100}

algo/tests/should-get-tests/large-r.should-get

-!LAUNCH: (for i in {1..100000}; do echo '>read' ; echo ccgtgtattactgtgcgagagagctgaatacttccagcactg ; done ;) > same-igh-100k.fa ; $LAUNCHER ../../vidjil -G ../../germline/IGH -r 5000  -w 15  same-igh-100k.fa; rm -f same-igh-100k.fa
+!LAUNCH: (for i in {1..100000}; do echo '>read' ; echo ccgtgtattactgtgcgagagagctgaatacttccagcactg ; done ;) > same-igh-100k.fa ; $LAUNCHER $VIDJIL_DIR/vidjil -G $VIDJIL_DIR/germline/IGH -r 5000  -w 15  same-igh-100k.fa; rm -f same-igh-100k.fa
 
 $ Find a unique clone with all reads
 1:>clone-001--IGH--0100000--100.--window

algo/tests/should-get-tests/multi-complete.should-get

-!LAUNCH: ../../vidjil -g ../../germline ../../data/multi-complete.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -g $VIDJIL_DIR/germline $VIDJIL_DIR/data/multi-complete.fa
 
 $ Segment all the seven reads
 1:junction detected in 7 reads

algo/tests/should-get-tests/multi-i-complete.should-get

-!LAUNCH: ../../vidjil -g ../../germline -i ../../data/multi-complete.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -g $VIDJIL_DIR/germline -i $VIDJIL_DIR/data/multi-complete.fa
 
 $ Segment all the seven reads
 1:junction detected in 7 reads

algo/tests/should-get-tests/multi-i-short.should-get

-!LAUNCH: ../../vidjil -g ../../germline -i ../../data/multi-short.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -g $VIDJIL_DIR/germline -i $VIDJIL_DIR/data/multi-short.fa
 
 $ Segment all the seven reads
 1:junction detected in 7 reads

algo/tests/should-get-tests/multi-short-affects.should-get

-!LAUNCH: ../../vidjil -r 5 -K -g ../../germline ../../data/multi-short.fa ; head -n 17 out/multi-short.affects
+!LAUNCH: $VIDJIL_DIR/vidjil -r 5 -K -g $VIDJIL_DIR/germline $VIDJIL_DIR/data/multi-short.fa ; head -n 17 out/multi-short.affects
 # Testing .affects output (-K)
 
 $ First sequence (TRA), display sequence

algo/tests/should-get-tests/multi-short.should-get

-!LAUNCH: ../../vidjil -g ../../germline ../../data/multi-short.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -g $VIDJIL_DIR/germline $VIDJIL_DIR/data/multi-short.fa
 
 $ Segment all the seven reads
 1:junction detected in 7 reads

algo/tests/should-get-tests/multi-tiny-segment.should-get

-!LAUNCH: ../../vidjil -c segment -g ../../germline ../../data/multi-tiny.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -c segment -g $VIDJIL_DIR/germline $VIDJIL_DIR/data/multi-tiny.fa
 
 $ Do not segment any of the seven reads
 7:UNSEG

algo/tests/should-get-tests/no_representative.should-get

-!LAUNCH: ../../vidjil -r 1 -k 4 -w 20 -z 0 -c clones -V ../../data/toy_V.fa -J ../../data/toy_J.fa ../../data/no_representative.fa 
+!LAUNCH: $VIDJIL_DIR/vidjil -r 1 -k 4 -w 20 -z 0 -c clones -V $VIDJIL_DIR/data/toy_V.fa -J $VIDJIL_DIR/data/toy_J.fa $VIDJIL_DIR/data/no_representative.fa 
 
 $ Short reads properly segmented
 1:SEG_+.* -> .* 4

algo/tests/should-get-tests/overlap-stdout.should-get

-!LAUNCH: ../../vidjil -c segment -G ../../germline/IGH -A ../../data/overlap-d-j.fa | tail -4 | tr -d '\\\\n' | wc -c
+!LAUNCH: $VIDJIL_DIR/vidjil -c segment -G $VIDJIL_DIR/germline/IGH -A $VIDJIL_DIR/data/overlap-d-j.fa | tail -4 | tr -d '\\\\n' | wc -c
 
 $ Exported sequence has all the bases
 1:116

algo/tests/should-get-tests/representative-N.should-get

-!LAUNCH: ../../vidjil -A -G ../../germline/TRG ./should-vdj-tests/ext-nucleotides-N.should-vdj.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -A -G $VIDJIL_DIR/germline/TRG ./should-vdj-tests/ext-nucleotides-N.should-vdj.fa
 
 $ Segments on TRG
 1: TRG .* -> .* 1

algo/tests/should-get-tests/revcomp.should-get

-!LAUNCH: $LAUNCHER ../../vidjil -k 9 -G ../../germline/IGH  -K -c clones ../../data/revcomp.fa ; grep 'X.X.X' out/revcomp.affects | sed 's/[^X]//g'  | sort -u ; grep '#>'  out/revcomp.affects | sed 's/.*SEG.../e-value:/' | cut -f 1 -d' '
+!LAUNCH: $LAUNCHER $VIDJIL_DIR/vidjil -k 9 -G $VIDJIL_DIR/germline/IGH  -K -c clones $VIDJIL_DIR/data/revcomp.fa ; grep 'X.X.X' out/revcomp.affects | sed 's/[^X]//g'  | sort -u ; grep '#>'  out/revcomp.affects | sed 's/.*SEG.../e-value:/' | cut -f 1 -d' '
 
 $ Segments both reads, normal and reverse
 1:junction detected in 2 reads

algo/tests/should-get-tests/segment_S22.should-get

-!LAUNCH: ../../vidjil -G ../../germline/IGH -c segment ../../data/segment_S22.fa | grep '^>' 
+!LAUNCH: $VIDJIL_DIR/vidjil -G $VIDJIL_DIR/germline/IGH -c segment $VIDJIL_DIR/data/segment_S22.fa | grep '^>' 
 
 $ First sequence Stanford
 #                            164        175                 195     203

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

-!LAUNCH: ../../vidjil -v -g ../../germline -c segment ../../data/segment_simul.fa | grep '^[>#]'
+!LAUNCH: $VIDJIL_DIR/vidjil -v -g $VIDJIL_DIR/germline -c segment $VIDJIL_DIR/data/segment_simul.fa | grep '^[>#]'
 
 
 $ First sequence, easy segmentation (no error, few deletions at the windows, small N)

algo/tests/should-get-tests/stanford-X100-sampled.should-get

-!LAUNCH: ../../vidjil -y 0 -X 100 -G ../../germline/IGH ../../data/Stanford_S22.fasta
+!LAUNCH: $VIDJIL_DIR/vidjil -y 0 -X 100 -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/Stanford_S22.fasta
 
 $ Skip the good number of reads
 1:Processing every 131th read

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

-!LAUNCH: ../../vidjil -\# FA -k 16 -z 0 -w 60 -r 5 -o out2 -u -U -v -G ../../germline/IGH ../../data/Stanford_S22.fasta ; tail out2/Stanford_S22.segmented.vdj.fa ; grep UNSEG out2/Stanford_S22.unsegmented.vdj.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -\# FA -k 16 -z 0 -w 60 -r 5 -o out2 -u -U -v -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/Stanford_S22.fasta ; tail out2/Stanford_S22.segmented.vdj.fa ; grep UNSEG out2/Stanford_S22.unsegmented.vdj.fa
 # Testing uncommon and debug options 
 
 $ verbose (-v)

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

-!LAUNCH: ../../vidjil -z 2 -r 5 -a -G ../../germline/IGH ../../data/Stanford_S22.fasta ; cat out/seq/clone.fa-2
+!LAUNCH: $VIDJIL_DIR/vidjil -z 2 -r 5 -a -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/Stanford_S22.fasta ; cat out/seq/clone.fa-2
 # Testing detailed clone output (-a)
 
 $ Detailed clone output (out/seq/clone.fa-2), germline

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

-!REQUIRES: python ../../tools/check_python_version.py
-!LAUNCH: ../../vidjil -z 0 -w 60 -G ../../germline/IGH ../../data/Stanford_S22.fasta ; python ../../tools/fuse.py out/Stanford_S22.vidjil out/Stanford_S22.vidjil -o out/fused.data ; cat out/fused.data | python ../../tools/format_json.py -1
+!REQUIRES: python $VIDJIL_DIR/tools/check_python_version.py
+!LAUNCH: $VIDJIL_DIR/vidjil -z 0 -w 60 -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/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
 e1:"original_names": ["../../data/Stanford_S22.fasta", "../../data/Stanford_S22.fasta"]

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

-!LAUNCH: (cd ../.. ; $LAUNCHER ./vidjil -c germlines -t 100 -s '######-######' data/Stanford_S22.fasta)
+!LAUNCH: (cd $VIDJIL_DIR ; $LAUNCHER ./vidjil -c germlines -t 100 -s '######-######' data/Stanford_S22.fasta)
 
 $ number of reads and kmers
 1:13153 reads, 3020179 kmers

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

-!REQUIRES: python ../../tools/check_python_version.py
-!LAUNCH: ../../vidjil -z 1 -G ../../germline/IGH -w 60 -r 5 -e 10 -b data  ../../data/Stanford_S22.fasta > /dev/null ; cat out/data.vidjil | python ../../tools/format_json.py -1
+!REQUIRES: python $VIDJIL_DIR/tools/check_python_version.py
+!LAUNCH: $VIDJIL_DIR/vidjil -z 1 -G $VIDJIL_DIR/germline/IGH -w 60 -r 5 -e 10 -b data  $VIDJIL_DIR/data/Stanford_S22.fasta > /dev/null ; cat out/data.vidjil | python $VIDJIL_DIR/tools/format_json.py -1
 
 $ Number of reads
 e1:"total": [13153]

algo/tests/should-get-tests/stanford-k14.should-get

-!LAUNCH: ../../vidjil -y 0 -k 14 -G ../../germline/IGH ../../data/Stanford_S22.fasta
+!LAUNCH: $VIDJIL_DIR/vidjil -y 0 -k 14 -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/Stanford_S22.fasta
 !LOG: stanford-k14.log
 
 $ Find the good number of windows in Stanford S22 (contiguous seed 14)

algo/tests/should-get-tests/stanford-labels-FaW.should-get

-!LAUNCH: ../../vidjil -t 0 -G ../../germline/IGH -FaW GAGAGGTTACTATGATAGTAGTGGTTATTACGGGGTAGGGCAGTACTACT ../../data/Stanford_S22.fasta ; cat out/seq/clone.fa-1
+!LAUNCH: $VIDJIL_DIR/vidjil -t 0 -G $VIDJIL_DIR/germline/IGH -FaW GAGAGGTTACTATGATAGTAGTGGTTATTACGGGGTAGGGCAGTACTACT $VIDJIL_DIR/data/Stanford_S22.fasta ; cat out/seq/clone.fa-1
 
 $ Keep only one windows, the one given by -W, with only 5 reads (it is actually the second clone in Stanford_S22.fasta)
 1: keep 1 windows in 5 reads

algo/tests/should-get-tests/stanford-labels-W.should-get

-!LAUNCH: ../../vidjil -z 0 -G ../../germline/IGH -r 5 -W GAGAGATGGACGGGATACGTAAAACGACATATGGTTCGGGGTTTGGTGCT ../../data/Stanford_S22.fasta
+!LAUNCH: $VIDJIL_DIR/vidjil -z 0 -G $VIDJIL_DIR/germline/IGH -r 5 -W GAGAGATGGACGGGATACGTAAAACGACATATGGTTCGGGGTTTGGTGCT $VIDJIL_DIR/data/Stanford_S22.fasta
 
 $ Some clone has only one read, bypassing the -r 5 option, and the good label
 1: clone-00..*0001-.* -W

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

-!LAUNCH: ../../vidjil -z 0 -G ../../germline/IGH -r 5 -l ../../data/Stanford_S22.label ../../data/Stanford_S22.fasta
+!LAUNCH: $VIDJIL_DIR/vidjil -z 0 -G $VIDJIL_DIR/germline/IGH -r 5 -l $VIDJIL_DIR/data/Stanford_S22.label $VIDJIL_DIR/data/Stanford_S22.fasta
 
 $ Some clone has only one read, bypassing the -r 5 option, and the good label
 1: clone-00..*0001-.* my-clone

algo/tests/should-get-tests/stanford-segment-x100-first.should-get

-!LAUNCH: ../../vidjil -c segment -x 2 -G ../../germline/IGH ../../data/Stanford_S22.fasta
+!LAUNCH: $VIDJIL_DIR/vidjil -c segment -x 2 -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/Stanford_S22.fasta
 
 $ Segments the good number of sequences in Stanford S22 
 2: >lcl

algo/tests/should-get-tests/stanford-w100.should-get

-!LAUNCH: ../../vidjil -e 10 -y 0 -s '#####-#####' -w 100 -G ../../germline/IGH ../../data/Stanford_S22.fasta
+!LAUNCH: $VIDJIL_DIR/vidjil -e 10 -y 0 -s '#####-#####' -w 100 -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/Stanford_S22.fasta
 !LOG: stanford-w100.log
 
 $ Find the good number of "too short sequences" for windows of size 100

algo/tests/should-get-tests/stanford-x100-first.should-get

-!LAUNCH: ../../vidjil -y 0 -x 100 -G ../../germline/IGH ../../data/Stanford_S22.fasta
+!LAUNCH: $VIDJIL_DIR/vidjil -y 0 -x 100 -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/Stanford_S22.fasta
 
 $ Analyze the good number of sequences in Stanford S22 
 1: found 99 ..-windows in 100 reads .100. of 100 reads

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

-!LAUNCH: ../../vidjil -e 10 -z 0 -V ../../germline/IGHV.fa -D ../../germline/IGHD.fa -J ../../germline/IGHJ.fa -s \\\\#\\\\#\\\\#\\\\#\\\\#\\\\#-\\\\#\\\\#\\\\#\\\\#\\\\#\\\\# ../../data/Stanford_S22.fasta
+!LAUNCH: $VIDJIL_DIR/vidjil -e 10 -z 0 -V $VIDJIL_DIR/germline/IGHV.fa -D $VIDJIL_DIR/germline/IGHD.fa -J $VIDJIL_DIR/germline/IGHJ.fa -s \\\\#\\\\#\\\\#\\\\#\\\\#\\\\#-\\\\#\\\\#\\\\#\\\\#\\\\#\\\\# $VIDJIL_DIR/data/Stanford_S22.fasta
 
 $ Parses IGHV.fa germline
 1: 102221 bp in 350 sequences

algo/tests/should-get-tests/trb-only-VJ.should-get

-!LAUNCH: ../../vidjil -A -G ../../germline/TRB ../../data/trb-only-VJ.fa ; cat out/trb-only-VJ.vidjil
+!LAUNCH: $VIDJIL_DIR/vidjil -A -G $VIDJIL_DIR/germline/TRB $VIDJIL_DIR/data/trb-only-VJ.fa ; cat out/trb-only-VJ.vidjil
 
 $ Segments the read on TRB (the information is given twice, stdout + .vidjil)
 2: TRB .* -> .* 1

algo/tests/should-get-tests/trd-dd2-dd3-complete.should-get

-!LAUNCH: ../../vidjil -A -g ../../germline ../../data/trd-dd2-dd3.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -A -g $VIDJIL_DIR/germline $VIDJIL_DIR/data/trd-dd2-dd3.fa
 
 $ Segment only 2 reads, because there is no -i
 1:junction detected in 2 reads .40..

algo/tests/should-get-tests/trd-dd2-dd3-incomplete.should-get

-!LAUNCH: ../../vidjil -w 10 -e 10 -A -i -g ../../germline ../../data/trd-dd2-dd3.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -w 10 -e 10 -A -i -g $VIDJIL_DIR/germline $VIDJIL_DIR/data/trd-dd2-dd3.fa
 
 $ Segment all 5 reads, thanks to -i
 1:junction detected in 5 reads .100..

algo/tests/should-get-tests/trd-dd2-dd3.should-get

-!LAUNCH: ../../vidjil -e 10 -k 8 -w 20 -V ../../germline/TRDV.fa -V ../../germline/TRDD2_upstream.fa -J ../../germline/TRDD3_downstream.fa -J ../../germline/TRDJ.fa ../../data/trd-dd2-dd3.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -e 10 -k 8 -w 20 -V $VIDJIL_DIR/germline/TRDV.fa -V $VIDJIL_DIR/germline/TRDD2_upstream.fa -J $VIDJIL_DIR/germline/TRDD3_downstream.fa -J $VIDJIL_DIR/germline/TRDJ.fa $VIDJIL_DIR/data/trd-dd2-dd3.fa
 
 $ Segment all 5 reads, thanks to TRDD2 and TRDD3
 1: junction detected in 5 reads .100..

algo/tests/should-get-tests/unsegmentation-causes.should-get

-!LAUNCH: ../../vidjil -g ../../germline -u ../../data/segmentation-2.fa ; cat out/segmentation-2.unsegmented.vdj.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -g $VIDJIL_DIR/germline -u $VIDJIL_DIR/data/segmentation-2.fa ; cat out/segmentation-2.unsegmented.vdj.fa
 
 $ Only one sequence is segmented, but it is too small for a window (too short w)
 1: junction detected in 1 reads

algo/tests/should-get-tests/vidjil-error-no-germline.should-get

-!LAUNCH: ../../vidjil 2>&1
+!LAUNCH: $VIDJIL_DIR/vidjil 2>&1
 !EXIT_CODE: 1
 
 $ Error, no germlines

algo/tests/should-get-tests/vidjil-error-w-small.should-get

-!LAUNCH: ../../vidjil -w 10 -G ../../germline/IGH ../../data/Stanford-S22.fa 2>&1
+!LAUNCH: $VIDJIL_DIR/vidjil -w 10 -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/Stanford-S22.fa 2>&1
 !EXIT_CODE: 1
 
 $ Error, too small -w

algo/tests/should-get-tests/vidjil-help.should-get

-!LAUNCH: ../../vidjil -h  2> out-v ; cat out-v
+!LAUNCH: $VIDJIL_DIR/vidjil -h  2> out-v ; cat out-v
 
 $ License
 1:Vidjil is free software

algo/tests/should-get-tests/vidjil-warning-A.should-get

-!LAUNCH: ../../vidjil -A -G ../../germline/IGH ../../data/toy_V.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -A -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/toy_V.fa
 
 $ Warning, -A
 1:WARNING

algo/tests/should-get-tests/vidjil-warning-z.should-get

 !NO_LAUNCHER:
-!LAUNCH: ../../vidjil -z 200 -G ../../germline/IGH ../../data/trd-dd2-dd3.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -z 200 -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/trd-dd2-dd3.fa
 
 $ Warning, -z
 1:WARNING

algo/tests/should-get-tests/vidjil_long_segment.should-get

-!LAUNCH: ../../vidjil -G ../../germline/IGH ../../data/long-segmentation.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -G $VIDJIL_DIR/germline/IGH $VIDJIL_DIR/data/long-segmentation.fa
 
 $ Sequence should be segmented by k-mer segmenter
 e1:SEG_+ -> 1 

algo/tests/should-get-tests/vidjil_revcomp_VdJa.should-get

-!LAUNCH: ../../vidjil -i -g ../../germline/ -r 1 -c clones ../../data/revcomp-VdJa.fa
+!LAUNCH: $VIDJIL_DIR/vidjil -i -g $VIDJIL_DIR/germline/ -r 1 -c clones $VIDJIL_DIR/data/revcomp-VdJa.fa
 
 $ Just one window found
 1:==> found 1 ..-windows

algo/tests/should-get-tests/vidjil_revcomp_s22.should-get

-!LAUNCH: $LAUNCHER ../../vidjil -z 1 -k 9 -G ../../germline/IGH -% 0.001 -r 2 -x 1000 -y 1 -c clones ../../data/Stanford_S22.fasta | sed 's/--IGH--.*VDJ\\(.*\\).$/\\1/' | sed 's/IGH SEG_./IGH SEG_X/' > vidjil_s22.log && $LAUNCHER ../../vidjil -z 1 -k 9 -G ../../germline/IGH -% 0.001 -r 2 -x 1000 -y 1 -c clones ../../data/Stanford_S22.rc.fasta | sed 's/--IGH--.*VDJ\\(.*\\).$/\\1/' |  sed 's/IGH SEG_./IGH SEG_X/' > vidjil_s22_rc.log && diff out/Stanford_S22{,.rc}.vidjil | grep GGG && diff vidjil_s22.log vidjil_s22_rc.log
+!LAUNCH: $LAUNCHER $VIDJIL_DIR/vidjil -z 1 -k 9 -G $VIDJIL_DIR/germline/IGH -% 0.001 -r 2 -x 1000 -y 1 -c clones $VIDJIL_DIR/data/Stanford_S22.fasta | sed 's/--IGH--.*VDJ\\(.*\\).$/\\1/' | sed 's/IGH SEG_./IGH SEG_X/' > vidjil_s22.log && $LAUNCHER $VIDJIL_DIR/vidjil -z 1 -k 9 -G $VIDJIL_DIR/germline/IGH -% 0.001 -r 2 -x 1000 -y 1 -c clones $VIDJIL_DIR/data/Stanford_S22.rc.fasta | sed 's/--IGH--.*VDJ\\(.*\\).$/\\1/' |  sed 's/IGH SEG_./IGH SEG_X/' > vidjil_s22_rc.log && diff out/Stanford_S22{,.rc}.vidjil | grep GGG && diff vidjil_s22.log vidjil_s22_rc.log
 !EXIT_CODE: 1
 
 $ Same number segmented