Update htqb docs #11, #5

docs/source/hqtb/about-pipeline.rst

@@ -19,13 +19,36 @@ The pipeline consists of five main steps:
 
 .. toctree::
     :maxdepth: 3
-    :numbered:
 
-    Bidirectional BLAST <step-desc/bidirect_blast.rst>
-    Draft Model Generation <step-desc/gen_draft.rst>
-    Model Refinement <step-desc/refinement.rst>
-    Validation <step-desc/validation.rst>
-    Analysis <step-desc/analysis.rst>
+    Step 1: Bidirectional BLAST <step-desc/bidirect_blast.rst>
+    Step 2: Draft Model Generation <step-desc/gen_draft.rst>
+
+- | Step 3: Model Refinement 
+  | Refine the previously generated draft model to make the model more complete and strain-specific. 
+  In other words, fitting it more closely to the input genome.
+
+    - :doc:`Part 1: Extension <step-desc/refine-parts/extension>`
+    - :doc:`Part 2: Clean-up <step-desc/refine-parts/cleanup>`
+    - :doc:`Part 3: Annotation <step-desc/refine-parts/annot>`
+    - :doc:`Part 4: Smoothing <step-desc/refine-parts/smooth>`
+
+.. toctree::
+    :maxdepth: 3
+
+    Step 4: Validation <step-desc/validation.rst>
+    Step 5: Analysis <step-desc/analysis.rst>
+
+
+.. toctree::
+    :hidden:
+    :maxdepth: 2
+
+    Part 1: Extension <step-desc/refine-parts/extension.rst>
+    Part 2: Clean-up <step-desc/refine-parts/cleanup.rst>
+    Part 3: Annotation <step-desc/refine-parts/annot.rst>
+    Part 4: Smoothing <step-desc/refine-parts/smooth.rst>
+
+
 
 The wrapper function allows the curation of multiple models sequentially using the same 
 boudary parameters.

docs/source/hqtb/step-desc/gen_draft.rst

@@ -6,20 +6,22 @@ template model.
 
 .. note:: 
 
-    For the pipeline to work, the template model needs to be constructed from the genome that was
-    used as a reference during the alignment in step 1.
+    When running this step outside the context of the pipeline, the input files 
+    need to be related to each other. The identifiers of the genes 
+    for the matches of the bidirectional blast results and the genes in the template model
+    are the same or can be adjusted using the given parameters as needed.
 
 A graphical representation of the steps can be found below:
 
 .. image:: ../../images/modules/2_gendraft.png
 
 The draft model is constructed based on the idea of :footcite:t:`norsigian2020workflow` by:
 
-- Filtering the BLAST best bidirectional hits by their percentage identity value (PID > threshold suggests, a homolog was found).
+- Filtering the BLAST best bidirectional hits by their percentage identity value (only matches with a PID greater than a given threshold are considered homologous).
 - Choosing a medium for the new model. Options include:
 
     - Using the one from the template
-    - Setting all exchanges to open (not advised)
+    - Setting all exchanges to open (not advised, quickly leads to unrealistic growth rates)
     - Loading one from the ``refineGEMs`` database
 
 - Removing genes from the template that have no (found) homolog in the template genome.

docs/source/hqtb/step-desc/refinement.rst

@@ -9,7 +9,7 @@ The model refinement process is split into four parts:
 .. toctree::
     :maxdepth: 2
 
-    Extension <refine-parts/extension.rst>
-    Clean-up <refine-parts/cleanup.rst>
-    Annotation <refine-parts/annot.rst>
-    Smoothing <refine-parts/smooth.rst>
+    Part 1: Extension <refine-parts/extension.rst>
+    Part 2: Clean-up <refine-parts/cleanup.rst>
+    Part 3: Annotation <refine-parts/annot.rst>
+    Part 4: Smoothing <refine-parts/smooth.rst>