changed learnapp to learn_apply in all config and files that call config params

Author: jjacobson95

.test/config_learnapp.yaml

@@ -43,7 +43,7 @@ basis_dir: "/path/to/output/kmerize/"
 score_dir: "/path/to/output/scoring/"
 
 
-learnapp:
+learn_apply:
   save_apply_associations: False
   fragmentation: False
   version: absolute

resources/LA_config.yaml

@@ -42,8 +42,8 @@ model_dir: "/path/to/output/model/"
 basis_dir: "/path/to/output/kmerize/"
 score_dir: "/path/to/output/scoring/"
 
-# learnapp params 
-learnapp:
+# learn_apply params 
+learn_apply:
   save_apply_associations: False       # Bool, use True if needing to extract more details. Increases storage requirements substantially.
   fragmentation: True                  # Bool, fragments the data according to the variables below (version, frag_length, min_length, location, seed)
   version: absolute

resources/kmer-association-example/config.yaml

@@ -42,8 +42,8 @@ model_dir: "/path/to/output/model/"
 basis_dir: "/path/to/output/kmerize/"
 score_dir: "/path/to/output/scoring/"
 
-# learnapp params 
-learnapp:
+# learn_apply params 
+learn_apply:
   save_apply_associations: False       # Bool, use True if needing to extract more details. Increases storage requirements substantially.
   fragmentation: True                  # Bool, fragments the data according to the variables below (version, frag_length, min_length, location, seed)
   version: absolute

snekmer/rules/apply.smk

@@ -72,10 +72,10 @@ out_dir = skm.io.define_output_dir(
     config["alphabet"], config["k"], nested=config["nested_output"]
 )
 
-threshold_type = config["learnapp"]["threshold"]
-selection_type = config["learnapp"]["selection"]
+threshold_type = config["learn_apply"]["threshold"]
+selection_type = config["learn_apply"]["selection"]
 
-apply_cfg = config.get("learnapp", {})
+apply_cfg = config.get("learn_apply", {})
 concat_results = apply_cfg.get("apply_output", "snekmer_results.csv")
 extra_all = [concat_results] if concat_results is not None else []
 
@@ -101,7 +101,7 @@ rule all:
                 expand(
                     join(out_dir, "apply", "seq-annotation-scores-{nb}.csv"), nb=FAS
                 )
-                if config["learnapp"]["save_apply_associations"]
+                if config["learn_apply"]["save_apply_associations"]
                 else []
             )
         ],
@@ -129,12 +129,12 @@ rule apply:
         confidence_associations=expand("{conf}", conf=confidence_file),
         decoy_stats=expand("{decoy}", decoy=decoy_stats_file),
     params:
-        selection_type=config["learnapp"]["selection"],
-        threshold_type=config["learnapp"]["threshold"],
+        selection_type=config["learn_apply"]["selection"],
+        threshold_type=config["learn_apply"]["threshold"],
     output:
         seq_ann=(
             expand(join(out_dir, "apply", "seq-annotation-scores-{nb}.csv"), nb=FAS)
-            if config["learnapp"]["save_apply_associations"]
+            if config["learn_apply"]["save_apply_associations"]
             else []
         ),
         kmer_summary=join(out_dir, "apply", "kmer-summary-{nb}.csv"),

snekmer/rules/learn.smk

@@ -73,8 +73,8 @@ out_dir = skm.io.define_output_dir(
 )
 
 if (
-    config["learnapp"]["selection"] != "top_hit"
-    and config["learnapp"]["threshold"] == "None"
+    config["learn_apply"]["selection"] != "top_hit"
+    and config["learn_apply"]["threshold"] == "None"
 ):
     raise Exception(
         "The only selection method that allows for None is `top_hit`. Other methods inherently use a threshold"
@@ -99,11 +99,11 @@ rule all:
         join(out_dir, "learn", "kmer-counts-total.csv"),
         # Fragmentation outputs (only if enabled)
         expand(join(out_dir, "fragmented", "{nb}.fasta"), nb=FAS)
-        if config["learnapp"]["fragmentation"]
+        if config["learn_apply"]["fragmentation"]
         else [],
         # expand(join(out_dir, "vector_frag", "{nb}.npz"), nb=FAS)
         expand(join(out_dir, "vector", "vector_frag", "{nb}.npz"), nb=FAS)
-        if config["learnapp"]["fragmentation"]
+        if config["learn_apply"]["fragmentation"]
         else [],
         # Forward evaluation scores
         expand(
@@ -112,7 +112,7 @@ rule all:
                 "evaluate",
                 (
                     "eval_apply_sequences"
-                    if not config["learnapp"]["fragmentation"]
+                    if not config["learn_apply"]["fragmentation"]
                     else "eval_apply_frag"
                 ),
                 "seq-annotation-scores-{nb}.csv.gz",
@@ -126,7 +126,7 @@ rule all:
                 "evaluate",
                 (
                     "eval_apply_reversed"
-                    if not config["learnapp"]["fragmentation"]
+                    if not config["learn_apply"]["fragmentation"]
                     else "eval_apply_reversed_frag"
                 ),
                 "seq-annotation-scores-{nb}.csv.gz",
@@ -149,7 +149,7 @@ use rule unzip from process with:
         zipped=join(input_dir, "zipped", "{uz}.gz"),
 
 
-if config["learnapp"]["fragmentation"]:
+if config["learn_apply"]["fragmentation"]:
 
     rule fragmentation:
         input:
@@ -161,11 +161,11 @@ if config["learnapp"]["fragmentation"]:
         message:
             "Fragmenting sequences in {input.fasta}. Output written to {output.fasta_out}."
         params:
-            version=config["learnapp"]["version"],
-            frag_length=config["learnapp"]["frag_length"],
-            location=config["learnapp"]["location"],
-            min_length=config["learnapp"]["min_length"],
-            seed=config["learnapp"]["seed"],
+            version=config["learn_apply"]["version"],
+            frag_length=config["learn_apply"]["frag_length"],
+            location=config["learn_apply"]["location"],
+            min_length=config["learn_apply"]["min_length"],
+            seed=config["learn_apply"]["seed"],
         script:
             resource_path("snekmer", "scripts", "learn_fragment.py")
 
@@ -224,7 +224,7 @@ rule eval_apply_reverse_seqs:
         data=join(
             out_dir,
             "vector",
-            ("vector" if not config["learnapp"]["fragmentation"] else "vector_frag"),
+            ("vector" if not config["learn_apply"]["fragmentation"] else "vector_frag"),
             "{nb}.npz",
         ),
         annotation=expand("{an}", an=annot_files),
@@ -235,7 +235,7 @@ rule eval_apply_reverse_seqs:
             "evaluate",
             (
                 "eval_apply_reversed"
-                if not config["learnapp"]["fragmentation"]
+                if not config["learn_apply"]["fragmentation"]
                 else "eval_apply_reversed_frag"
             ),
             "seq-annotation-scores-{nb}.csv.gz",
@@ -254,7 +254,7 @@ rule reverse_decoy_evaluations:
                 "evaluate",
                 (
                     "eval_apply_reversed"
-                    if not config["learnapp"]["fragmentation"]
+                    if not config["learn_apply"]["fragmentation"]
                     else "eval_apply_reversed_frag"
                 ),
                 "seq-annotation-scores-{nb}.csv.gz",
@@ -276,7 +276,7 @@ rule eval_apply_sequences:
         data=join(
             out_dir,
             "vector",
-            ("vector" if not config["learnapp"]["fragmentation"] else "vector_frag"),
+            ("vector" if not config["learn_apply"]["fragmentation"] else "vector_frag"),
             "{nb}.npz",
         ),
         annotation=expand("{an}", an=annot_files),
@@ -287,7 +287,7 @@ rule eval_apply_sequences:
             "evaluate",
             (
                 "eval_apply_sequences"
-                if not config["learnapp"]["fragmentation"]
+                if not config["learn_apply"]["fragmentation"]
                 else "eval_apply_frag"
             ),
             "seq-annotation-scores-{nb}.csv.gz",
@@ -306,7 +306,7 @@ rule evaluate:
                 "evaluate",
                 (
                     "eval_apply_sequences"
-                    if not config["learnapp"]["fragmentation"]
+                    if not config["learn_apply"]["fragmentation"]
                     else "eval_apply_frag"
                 ),
                 "seq-annotation-scores-{nb}.csv.gz",
@@ -320,7 +320,7 @@ rule evaluate:
     message:
         "Calculating global confidence scores based on Apply results. Output written to {output.eval_glob}."
     params:
-        modifier=config["learnapp"]["conf_weight_modifier"],
+        modifier=config["learn_apply"]["conf_weight_modifier"],
     script:
         resource_path("snekmer", "scripts", "learn_evaluate_sequences.py")
 

snekmer/scripts/apply.py

@@ -14,9 +14,9 @@
 # Run script
 # ---------------------------------------------------------
 
-weight_top = config["learnapp"].get("weight_top", 0.5)
-weight_distance = config["learnapp"].get("weight_distance", 0.5)
-save_apply_associations = config["learnapp"]["save_apply_associations"]
+weight_top = config["learn_apply"].get("weight_top", 0.5)
+weight_distance = config["learn_apply"].get("weight_distance", 0.5)
+save_apply_associations = config["learn_apply"]["save_apply_associations"]
 
 apply = KmerCompare(
     snakemake.input.compare_associations,

snekmer/scripts/learn_eval_apply_sequences.py

@@ -238,7 +238,7 @@ def execute_all(self, config: str) -> None:
         kmer_counts = self.construct_kmer_counts_dataframe()
         kmer_counts, self.kmer_count_totals = match_kmer_counts_format(kmer_counts,self.kmer_count_totals)
         final_matrix_with_scores = self.calculate_cosine_similarity(kmer_counts)
-        if not config["learnapp"]["save_apply_associations"]:
+        if not config["learn_apply"]["save_apply_associations"]:
             final_matrix_with_scores = self.filter_top_two_values(
                 final_matrix_with_scores
             )

snekmer/scripts/learn_evaluate_sequences.py

@@ -96,10 +96,10 @@ def read_and_transform_input_data(
             header=0,
             engine="c",
         )
-        threshold_type = config["learnapp"]["threshold"]
-        selection_method = config["learnapp"]["selection"]
-        weight_top = config["learnapp"].get("weight_top", 0.5)
-        weight_distance = config["learnapp"].get("weight_distance", 0.5)
+        threshold_type = config["learn_apply"]["threshold"]
+        selection_method = config["learn_apply"]["selection"]
+        weight_top = config["learn_apply"].get("weight_top", 0.5)
+        weight_distance = config["learn_apply"].get("weight_distance", 0.5)
 
         if threshold_type == "None":
             threshold_type = None