From fb015e0eb441f3613f488b43eace65c1a862cf3a Mon Sep 17 00:00:00 2001
From: Puja Trivedi <pujatrivedi@Pujas-MacBook-Pro.local>
Date: Mon, 11 Mar 2024 11:18:14 -0700
Subject: [PATCH] updated scripts/README.md with linkml_trimmer details

---
 bkbit/scripts/README.md | 33 ++++++++++++++++-----------------
 1 file changed, 16 insertions(+), 17 deletions(-)

diff --git a/bkbit/scripts/README.md b/bkbit/scripts/README.md
index a538162..f224d2a 100644
--- a/bkbit/scripts/README.md
+++ b/bkbit/scripts/README.md
@@ -1,29 +1,28 @@
-# gfftranslator.py
-
-gfftranslator.py is a Python script that generates GeneAnnotation objects from data stored in gff files.  
+# linkml_trimmer.py
 
+linkml_trimmer returns a trimmed version of a linkml model. 
 ## Usage
 
 ```python
-from bkbit.scripts.gfftranslator import gff_to_gene_annotation
+# Step 1: import YamlTrimmer
+from bkbit.scripts.linkml_trimmer import YamlTrimmer
 
-# input_fname is the name of the input csv file 
-# Note: example input data can be found on Allen Teams under Knowledge Graph files. "20230412_subset_genome_annotation.csv" 
-input_fname = 'XXX.csv'
+# Step 2: initialize YamlTrimmer Object with a linkml model 
+trimmed_model = YamlTrimmer(path_to_linkml_model)
 
-# data_dir is the directory path where the input csv file exists
-data_dir = ' XXX/XXX/'
+# Step 3: define the classes, slots, and enums that should be included in the trimmed model
+classes = [...] # List of classes to keep
+slots = [...] # List of slots to keep
+enums = [...] # List of enums to keep
 
-# output_dir is the directory path where all of the generated output files will be saved 
-# Note: if output_dir does not exist, gff_to_gene_annotation will create the directory
-output_dir = 'XXX/XXX/'
+# Step 4: call the trim_model function with the selected classes/slots/enums 
+# Note: only classes is a required parameter. slots and enums are optional 
+trimmed_model.trim_model(classes, slots, enums)
 
-gff_to_gene_annotation(input_fname, data_dir, output_dir)
+# Step 5: call the serialize function to produce trimmed linkml model 
+trimmed_model.serialize()
 ```
 
 ## Notes
 
-1. Input csv file  
-a. Each row in the csv file contains a url to the .gff file as well as additional attributes to describe the dataset. The csv file must contain the following columns: authority, label, taxon_local_unique_identifier, version, gene_identifier_prefix, url. 
-2. Generated files  
-a. For each .gff file 3 files will be generated: (i) The raw data downloaded from the url provided will be saved as a csv file in the 'data_dir' directory provided. (ii) The parsed and cleaned data will be saved as a csv file in the 'output_dir' directory provided. (iii) The initialized GeneAnnotation objects will be saved as a list of json dictionaries in a json file in the 'output_dir' directory provided. 
+1. To produce bican_biolink.yaml call trim_model with classes = ['gene', 'genome', 'organism taxon', 'thing with taxon', 'material sample', 'procedure', 'entity', 'activity', 'named thing']