Skip to content

Commit

Permalink
get_phylomarkers_fun_lib 2024-04-16
Browse files Browse the repository at this point in the history 
- prepended missing "$bindir"/ to several iqtree calls in run_ASTRAL and compute_ASTRALspTree_branch_lenghts
- improved print_start_time progress reporting from run_ASTRAL and compute_ASTRALspTree_branch_lenghts
  • Loading branch information
@vinuesa
vinuesa committed Apr 17, 2024
1 parent 944f089 commit 8dfdec1
Showing 1 changed file with 19 additions and 20 deletions.
39 changes: 19 additions & 20 deletions lib/get_phylomarkers_fun_lib
View file
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
#: VERSION: 2024-04-13
#: VERSION: 2024-04-16
#: LIBRARY: get_phylomarkers_fun_lib
# * functions called by run_get_phylomarkers_pipeline.sh
# * part of the GET_PHYLOMARKERS software package
Expand Down Expand Up @@ -520,14 +520,13 @@ function run_ASTRAL()
then
concat_aln=concat_cdnAlns.fnainf
lmsg=" # computing best-fitting model for $concat_aln using fixed user-tree wASTRAL species tree"
msg "$lmsg" PRGR LBLUE
iqtree -s "$concat_aln" -m MF+ASC+F+G+R*ASC*F*G*R -g "wastral_top${num_markers}geneTrees.sptree" \
--prefix wastral_usertree_concat_model_selection -nt "$num_threads" -mset "$dna_mset" -cmax "$cmax" --fast --quiet
print_start_time && msg "$lmsg" PRGR LBLUE
"$bindir"/iqtree -s "$concat_aln" -m MF+ASC+F+G+R*ASC*F*G*R -g "wastral_top${num_markers}geneTrees.sptree" \
--prefix wastral_usertree_concat_model_selection -nt "$num_threads" -mset "$dna_mset" -cmax "$cmax" --fast &> /dev/null

best_wastral_constrained_IQT_concat_model=''
best_wastral_constrained_IQT_concat_model=$(awk '/chosen according to BIC/{print $3}' wastral_usertree_concat_model_selection.log)

if [[ -n "$best_wastral_constrained_IQT_concat_model" ]]
best_wastral_constrained_IQT_concat_model=$(awk '/Best-fit model according to BIC:/{print $6}' wastral_usertree_concat_model_selection.iqtree)

if [ -n "$best_wastral_constrained_IQT_concat_model" ]
then
msg " >>> best fit model for $concat_aln using fixed user-tree wASTRAL species tree: $best_wastral_constrained_IQT_concat_model" PROG GREEN
else
Expand All @@ -537,16 +536,15 @@ function run_ASTRAL()
else
concat_aln=concat_protAlns.faainf
lmsg=" # computing best-fitting model for $concat_aln using fixed user-tree wASTRAL species tree"
msg "$lmsg" PRGR LBLUE
print_start_time && msg "$lmsg" PRGR LBLUE

# NOTE: mixture models + ASC not yet implemented in iqtree v2.3.1 (Mar 25, 2024)
iqtree -s "$concat_aln" -m MF+F+G+R -g "wastral_top${num_markers}geneTrees.sptree" \
--prefix wastral_usertree_concat_model_selection -nt "$num_threads" -mset "$prot_mset" -cmax 10 --fast --quiet
"$bindir"/iqtree -s "$concat_aln" -m MF+F+G+R -g "wastral_top${num_markers}geneTrees.sptree" \
--prefix wastral_usertree_concat_model_selection -nt "$num_threads" -mset "$prot_mset" -cmax 10 --fast &> /dev/null

best_wastral_constrained_IQT_concat_model=''
best_wastral_constrained_IQT_concat_model=$(awk '/chosen according to BIC/{print $3}' wastral_usertree_concat_model_selection.log)
best_wastral_constrained_IQT_concat_model=$(awk '/Best-fit model according to BIC:/{print $6}' wastral_usertree_concat_model_selection.iqtree)

if [[ -n "$best_wastral_constrained_IQT_concat_model" ]]
if [ -n "$best_wastral_constrained_IQT_concat_model" ]
then
msg " >>> best fit model for $concat_aln using fixed user-tree wASTRAL species tree: $best_wastral_constrained_IQT_concat_model" PROG GREEN
else
Expand All @@ -556,15 +554,16 @@ function run_ASTRAL()
fi

# 4. compute_ASTRALspTree_branch_lenghts using best_wastral-constrained_IQT_concat_model and DNA|PROT \$concat_aln
msg "# running compute_ASTRALspTree_branch_lenghts $concat_aln $best_wastral_constrained_IQT_concat_model wastral_top${no_top_markers}geneTrees.sptree $num_threads ..." PROGR LBLUE
compute_ASTRALspTree_branch_lenghts "$concat_aln" "$best_wastral_constrained_IQT_concat_model" wastral_top"${no_top_markers}"geneTrees.sptree "$num_threads"

# 5.1 compute the gene concordance factor (gCF) for each branch of the ASTRAL-IV species tree
# as the fraction of decisive gene trees concordant with this branch
# NOTE: the IQTree_constrained_by_ASTRALspTree_concord.cf.tree produced by IQT may not be visualizable with FigTree!!!
msg "# computing the gene concordance factor (gCF) for each branch of the ASTRALIV species tree ..." PROGR LBLUE
print_start_time && msg "# computing the gene concordance factor (gCF) for each branch of the ASTRALIV species tree ..." PROGR LBLUE
long_msg=" > iqtree -t astral4_top${num_markers}geneTrees.sptree --gcf alltrees.nwk --prefix astral4_concord &> /dev/null"
(( DEBUG > 0 )) && msg "$long_msg" DEBUG NC
iqtree -t astral4_top"${num_markers}"geneTrees.sptree --gcf alltrees.nwk --prefix astral4_concord &> /dev/null
"$bindir"/iqtree -t astral4_top"${num_markers}"geneTrees.sptree --gcf alltrees.nwk --prefix astral4_concord &> /dev/null
check_output astral4_concord.cf.tree "$parent_PID"

# 5.2 add labels to astral4_concord.cf.tree
Expand All @@ -578,10 +577,10 @@ function run_ASTRAL()
fi

# 5.3 compute the gene concordance factor (gCF) for each branch of the wASTRAL species tree
msg "# computing the gene concordance factor (gCF) for each branch of the wASTRAL species tree ..." PROGR LBLUE
print_start_time && msg "# computing the gene concordance factor (gCF) for each branch of the wASTRAL species tree ..." PROGR LBLUE
long_msg=" > iqtree -t wastral_top${num_markers}geneTrees.sptree --gcf alltrees.nwk --prefix wastral_concord &> /dev/null"
(( DEBUG > 0 )) && msg "$long_msg" DEBUG NC
iqtree -t wastral_top"${num_markers}"geneTrees.sptree --gcf alltrees.nwk --prefix wastral_concord &> /dev/null
"$bindir"/iqtree -t wastral_top"${num_markers}"geneTrees.sptree --gcf alltrees.nwk --prefix wastral_concord &> /dev/null
check_output wastral_concord.cf.tree "$parent_PID"

# 5.4 add labels to wastral_concord.cf.tree
Expand Down Expand Up @@ -655,10 +654,10 @@ function compute_ASTRALspTree_branch_lenghts()
[[ "$threads" == 'AUTO' ]] && threads=10 # see note above on but in IQ-Tree 2.2.0 <<< !!!!

# compute ML branch lenghts of wASTRAL species tree with IQT using the best model found for concat_aln
msg "# computing ML branch lenghts of wASTRAL species tree with IQT using the best model found for concat_aln ..." PROGR LBLUE
print_start_time && msg "# computing ML branch lenghts of wASTRAL species tree with IQT using the best model $model found for concat_aln ..." PROGR LBLUE
long_msg=" > iqtree -s $concat_aln -m $model -g $astral_spTree --prefix $prefix -T $threads &> /dev/null"
(( DEBUG > 0 )) && msg "$long_msg" DEBUG NC
iqtree -s "$concat_aln" -m "$model" -g "$astral_spTree" --prefix "$prefix" -T "$threads" &> /dev/null
"$bindir"/iqtree -s "$concat_aln" -m "$model" -g "$astral_spTree" --prefix "$prefix" -T "$threads" &> /dev/null

# the IQT implementaion in v2.2.0 seems to be buggy, as for the same data
# the previous call is sometimes not successful (particularly in the Travis environment)
Expand Down

0 comments on commit 8dfdec1

Please sign in to comment.