update utils.run_ete_script() and remove ete_path/--ete-path

don't need it since we're on python 3 now

bin/bcr-phylo-run.py

@@ -24,7 +24,6 @@
 from python.event import RecombinationEvent
 import python.paircluster as paircluster
 
-ete_path = os.getenv('HOME') + '/anaconda_ete/bin'
 bcr_phylo_path = os.getenv('PWD') + '/packages/bcr-phylo-benchmark'
 ig_or_tr = 'ig'
 
@@ -212,9 +211,9 @@ def run_bcr_phylo(naive_seq, outdir, ievent, uid_str_len=None, igcr=None):
 
     cfo = None
     if args.n_procs == 1:
-        utils.run_ete_script(cmd, ete_path, dryrun=args.dry_run)
+        utils.run_ete_script(cmd, dryrun=args.dry_run)
     else:
-        cmd, _ = utils.run_ete_script(cmd, ete_path, return_for_cmdfos=True, tmpdir=outdir, dryrun=args.dry_run)
+        cmd = utils.run_ete_script(cmd, return_for_cmdfos=True, dryrun=args.dry_run)
         cfo = {'cmd_str' : cmd, 'workdir' : outdir, 'outfname' : bcr_phylo_fasta_fname(outdir)}
     sys.stdout.flush()
     return cfo
@@ -348,7 +347,7 @@ def translate_duplicate_pids(mpair, dup_translations):
     kdfname, nwkfname = '%s/kd-vals.csv' % outdir, '%s/simu.nwk' % outdir
     if not utils.output_exists(args, kdfname, outlabel='kd/nwk conversion', offset=4):  # eh, don't really need to check for both kd and nwk file, chances of only one being missing are really small, and it'll just crash when it looks for it a couple lines later
         cmd = './bin/read-bcr-phylo-trees.py --pickle-tree-file %s/%s_lineage_tree.p --kdfile %s --newick-tree-file %s' % (outdir, args.extrastr, kdfname, nwkfname)
-        utils.run_ete_script(cmd, ete_path, debug=args.n_procs==1)
+        utils.run_ete_script(cmd, debug=args.n_procs==1)
     nodefo = read_kdvals(kdfname)
     dtree = treeutils.get_dendro_tree(treefname=nwkfname)
     seqfos = utils.read_fastx(bcr_phylo_fasta_fname(outdir))  # output mutated sequences from bcr-phylo

bin/gctree-run.py

@@ -167,7 +167,8 @@ def parse_output():
 def convert_pickle_tree():
     assert False  # doesn't work yet
     cmd = '%s/bin/read-bcr-phylo-trees.py --pickle-tree-file %s --newick-tree-file %s/tree.nwk' % (utils.get_partis_dir(), args.infname, args.outdir)
-    utils.run_ete_script(cmd, None, conda_path=args.condapath, conda_env='ete3', pyversion='3')
+    # assert False  # this also needs updating
+    # utils.run_ete_script(cmd, None, conda_path=args.condapath, conda_env='ete3', pyversion='3')
 
 # ----------------------------------------------------------------------------------------
 parser = argparse.ArgumentParser()

bin/partis

@@ -1380,7 +1380,6 @@ parent_args.append({'name' : '--no-partition-plots', 'kwargs' : {'action' : 'sto
 parent_args.append({'name' : '--only-csv-plots', 'kwargs' : {'action' : 'store_true', 'help' : 'skip writing actual image files, which can quite be slow, and only write the csv/yaml summaries (where implemented)'}})
 parent_args.append({'name' : '--make-per-gene-plots', 'kwargs' : {'action' : 'store_true', 'help' : 'in addition to plots aggregating over genes, write plots displaying info for each gene of, e.g., per position shm rate, deletion frequencies'}})
 parent_args.append({'name' : '--make-per-gene-per-base-plots', 'kwargs' : {'action' : 'store_true', 'help' : 'in addition to the plots made by --make-per-gene-plots, also make the per-gene, per-base plots (i.e. showing A->T vs A->G (this is quite slow, like a few seconds per gene plot).'}})
-parent_args.append({'name' : '--ete-path', 'kwargs' : {'default' : ('%s/anaconda_ete/bin' % os.getenv('HOME')) if os.getenv('HOME') is not None else None, 'help' : 'Set to the string \'None\' to turn off.'}})
 parent_args.append({'name' : '--linearham-dir', 'kwargs' : {'default' : ('%s/work/linearham' % os.getenv('HOME')) if os.getenv('HOME') is not None else None, 'help' : 'path to linearham main dir (necessary if you want to use linearham without docker)'}})
 parent_args.append({'name' : '--meta-info-to-emphasize', 'kwargs' : {'help' : 'Input meta info (or regular annotation) key to emphasize (highlight in red) in various plots, similar to --queries-to-include. Specify as comma-separated key-value pair, for instance \'timepoints,+8d\' would highlight all sequences with timepoint \'+8d\'. Can be any annotation key or input meta key. For now only supports one key-val pair, but in future should support colon-separated list.'}})
 parent_args.append({'name' : '--meta-info-key-to-color', 'kwargs' : {'help' : 'Like --meta-info-to-emphasize, except for this key we choose a different color for each value (only in the slug/joy plots at the moment).'}})

bin/read-gctree-output.py

@@ -113,12 +113,13 @@ def run_cmd(action):
         cmd += ' --min-selection-metric-cluster-size 3 --treefname %s/%s --plotdir %s --selection-metrics-to-calculate lbi:aa-lbi:cons-dist-aa:lbr:aa-lbr:lbf:aa-lbf' % (args.gctreedir, args.tree_basename, 'paired-outdir' if args.paired_loci else '%s/selection-metrics/plots'%args.outdir)
         cmd += ' --extra-daffy-metrics lbi:aa-lbi'
         cmd += ' --label-root-node'
-        cmd += ' --add-selection-metrics-to-outfname --use-droplet-id-for-combo-id --selection-metric-plot-cfg %s' % ':'.join(treeutils.default_plot_cfg + ['distr', 'tree-mut-stats'])
+        plt_cfg = treeutils.default_plot_cfg + ['distr', 'tree-mut-stats']
+        if args.no_tree_plots:
+            plt_cfg = [t for t in plt_cfg if t != 'tree']
+        cmd += ' --add-selection-metrics-to-outfname --use-droplet-id-for-combo-id --selection-metric-plot-cfg %s' % ':'.join(plt_cfg)
         if args.slice_bin_fname is not None:
             cmd += ' --slice-bin-fname %s' % args.slice_bin_fname
         cmd += ' --choose-all-abs --chosen-ab-fname %s/chosen-abs.csv' % args.outdir  #  --debug 1
-        if args.no_tree_plots:
-            cmd += ' --ete-path None'
     if args.n_procs is not None:
         cmd += ' --n-procs %d' % args.n_procs
     utils.simplerun(cmd, logfname='%s/%s.log'%(args.outdir, action), dryrun=args.dry)

bin/smetric-run.py

@@ -38,9 +38,8 @@
 # NOTE extra required args are set in treeutils plot_tree_metrics()
 args.cluster_indices = utils.get_arg_list(args.cluster_indices, intify_with_ranges=True)
 args.selection_metric_plot_cfg = utils.get_arg_list(args.selection_metric_plot_cfg, choices=treeutils.all_plot_cfg)
-ete_path, workdir = None, None
+workdir = None, None
 if args.make_tree_plots or 'tree' in args.selection_metric_plot_cfg:
-    ete_path = '%s/anaconda_ete/bin' % os.getenv('HOME')
     workdir = utils.choose_random_subdir('/tmp/%s/tree-metrics' % os.getenv('USER'))
 
 if args.n_max_queries != -1:
@@ -56,8 +55,8 @@
 if args.metric_method == 'dtr':
     raise Exception('I think the [new] first arg here (metrics_to_calc) isn\'t right, but don\'t want to test cause i don\'t care about dtr')
     treeutils.add_smetrics(args, ['lbi', 'lbr', 'dtr'], None, args.lb_tau, base_plotdir=args.base_plotdir,
-                           train_dtr=args.action=='train', dtr_cfg=args.dtr_cfg, true_lines_to_use=true_lines, ete_path=ete_path, workdir=workdir)  # NOTE if you need this in the future you may want to add tree_inference_method 'gctree' and tree_inference_outdir
+                           train_dtr=args.action=='train', dtr_cfg=args.dtr_cfg, true_lines_to_use=true_lines, workdir=workdir)  # NOTE if you need this in the future you may want to add tree_inference_method 'gctree' and tree_inference_outdir
 else:
     treeutils.calculate_individual_tree_metrics(args.metric_method, true_lines, base_plotdir=args.base_plotdir, lb_tau=args.lb_tau, only_csv=args.only_csv_plots,
                                                 min_cluster_size=args.min_selection_metric_cluster_size, include_relative_affy_plots=args.include_relative_affy_plots,
-                                                dont_normalize_lbi=args.dont_normalize_lbi, ete_path=ete_path, workdir=workdir, cluster_indices=args.cluster_indices, only_look_upwards=args.only_look_upwards, args=args) #, debug=True)
+                                                dont_normalize_lbi=args.dont_normalize_lbi, workdir=workdir, cluster_indices=args.cluster_indices, only_look_upwards=args.only_look_upwards, args=args) #, debug=True)

python/lbplotting.py

@@ -1544,7 +1544,7 @@ def ctypetitle(ct):
         add_fn(fnames, fn=fn)
 
 # ----------------------------------------------------------------------------------------
-def get_lb_tree_cmd(treestr, outfname, lb_metric, affy_key, ete_path, subworkdir, metafo=None, tree_style=None, queries_to_include=None, label_all_nodes=False, label_leaf_nodes=False, label_root_node=False, seq_len=None,
+def get_lb_tree_cmd(treestr, outfname, lb_metric, affy_key, subworkdir, metafo=None, tree_style=None, queries_to_include=None, label_all_nodes=False, label_leaf_nodes=False, label_root_node=False, seq_len=None,
                     meta_info_key_to_color=None, meta_info_to_emphasize=None, node_size_key=None, branch_color_key=None, uid_translations=None, node_label_regex=None):
     treefname = '%s/tree.nwk' % subworkdir
     metafname = '%s/meta.yaml' % subworkdir
@@ -1586,12 +1586,12 @@ def get_lb_tree_cmd(treestr, outfname, lb_metric, affy_key, ete_path, subworkdir
         cmdstr += ' --branch-color-key %s' % branch_color_key
     if node_label_regex is not None:
         cmdstr += ' --node-label-regex %s' % node_label_regex
-    cmdstr, _ = utils.run_ete_script(cmdstr, ete_path, return_for_cmdfos=True, tmpdir=subworkdir, extra_str='        ')
+    cmdstr = utils.run_ete_script(cmdstr, return_for_cmdfos=True, extra_str='        ')
 
     return {'cmd_str' : cmdstr, 'workdir' : subworkdir, 'outfname' : outfname, 'workfnames' : [treefname, metafname]}
 
 # ----------------------------------------------------------------------------------------
-def plot_lb_trees(args, metric_methods, baseplotdir, lines, ete_path, base_workdir, is_true_line=False, tree_style=None, fnames=None):
+def plot_lb_trees(args, metric_methods, baseplotdir, lines, base_workdir, is_true_line=False, tree_style=None, fnames=None):
     add_fn(fnames, new_row=True)
     workdir = '%s/ete3-plots' % base_workdir
     plotdir = baseplotdir + '/trees'
@@ -1614,7 +1614,7 @@ def plot_lb_trees(args, metric_methods, baseplotdir, lines, ete_path, base_workd
             if affy_key in line:  # either 'affinities' or 'relative_affinities'
                 metafo[utils.reversed_input_metafile_keys[affy_key]] = {uid : affy for uid, affy in zip(line['unique_ids'], line[affy_key])}
             outfname = '%s/%s-tree-iclust-%d%s.svg' % (plotdir, lb_metric, iclust, '-relative' if 'relative' in affy_key else '')
-            cmdfos += [get_lb_tree_cmd(treestr, outfname, lb_metric, affy_key, ete_path, '%s/sub-%d' % (workdir, len(cmdfos)), metafo=metafo, tree_style=tree_style, queries_to_include=qtis,
+            cmdfos += [get_lb_tree_cmd(treestr, outfname, lb_metric, affy_key, '%s/sub-%d' % (workdir, len(cmdfos)), metafo=metafo, tree_style=tree_style, queries_to_include=qtis,
                                        label_all_nodes=args.label_tree_nodes, label_leaf_nodes=args.label_leaf_nodes, label_root_node=args.label_root_node, uid_translations=altids, node_label_regex=args.node_label_regex,
                                        seq_len=float(numpy.mean([len(s) for s in line['seqs']])))]
             add_fn(fnames, fn=outfname, n_per_row=4)

python/partitiondriver.py

@@ -390,7 +390,7 @@ def calc_tree_metrics(self, annotation_dict, annotation_list=None, cpath=None):
             print('    --seed-unique-id: restricting selection metric calculation to seed cluster in best partition (mostly to avoid fasttree crash on duplicate uids)')
             annotation_dict = OrderedDict([(uidstr, line) for uidstr, line in annotation_dict.items() if self.args.seed_unique_id in line['unique_ids'] and line['unique_ids'] in cpath.partitions[cpath.i_best]])
         treeutils.add_smetrics(self.args, self.args.selection_metrics_to_calculate, annotation_dict, self.args.lb_tau, reco_info=self.reco_info,  # NOTE keys in <annotation_dict> may be out of sync with 'unique_ids' if we add inferred ancestral seqs here
-                               use_true_clusters=self.reco_info is not None, base_plotdir=self.args.plotdir, ete_path=self.args.ete_path, workdir=self.args.workdir,
+                               use_true_clusters=self.reco_info is not None, base_plotdir=self.args.plotdir, workdir=self.args.workdir,
                                outfname=self.args.selection_metric_fname, glfo=self.glfo, tree_inference_outdir=self.args.tree_inference_outdir, debug=self.args.debug)
 
     # ----------------------------------------------------------------------------------------

python/partitionplotter.py

@@ -798,7 +798,7 @@ def vmuts(vclass_muts, mutfo, uid):
             qtis = None if self.args.queries_to_include is None else [q for q in self.args.queries_to_include if q in annotation['unique_ids']]  # NOTE make sure to *not* modify args.queries_to_include
             altids = [(u, au) for u, au in zip(annotation['unique_ids'], annotation['alternate-uids']) if au is not None] if 'alternate-uids' in annotation else None
             mfo, cdr3fo = get_metafo(annotation, iclust)
-            cfo = lbplotting.get_lb_tree_cmd(self.get_treestr(iclust), '%s/%s.svg'%(plotdir, plotname), None, None, self.args.ete_path, '%s/sub-%d'%(workdir, len(cmdfos)), metafo=mfo,
+            cfo = lbplotting.get_lb_tree_cmd(self.get_treestr(iclust), '%s/%s.svg'%(plotdir, plotname), None, None, '%s/sub-%d'%(workdir, len(cmdfos)), metafo=mfo,
                                              queries_to_include=qtis, meta_info_key_to_color=self.args.meta_info_key_to_color, meta_info_to_emphasize=self.args.meta_info_to_emphasize, uid_translations=altids,
                                              label_all_nodes=self.args.label_tree_nodes, label_leaf_nodes=self.args.label_leaf_nodes, label_root_node=self.args.label_root_node, node_size_key=self.args.node_size_key, branch_color_key=self.args.branch_color_key, node_label_regex=self.args.node_label_regex)
             cmdfos.append(cfo)

python/processargs.py

@@ -583,6 +583,3 @@ def process_corr_values(cvals, estr=''):
             raise Exception('have to specify --sw-cachefname or --parameter-dir, since we need sw info to calculate linearham inputs')
         if args.extra_annotation_columns is None or 'linearham-info' not in args.extra_annotation_columns:
             args.extra_annotation_columns = utils.add_lists(args.extra_annotation_columns, ['linearham-info'])
-
-    if args.ete_path is not None and args.ete_path == 'None':  # it's nice to be able to unset this from the command line (so we don't make the slow tree plots)
-        args.ete_path = None