Added FAQs section

README.md

@@ -42,35 +42,42 @@ Full CLI options (check out with ``geneplexus --help``)
 ```txt
 Run the GenePlexus pipline on a input gene list.
 
-optional arguments:
+options:
   -h, --help            show this help message and exit
-  -i , --input_file     Input gene list (.txt) file (one gene per line). (default: None)
+  -i , --input_file     Input gene list (.txt) file. (default: None)
   -d , --gene_list_delimiter
                         Delimiter used in the gene list. Use 'newline' if the genes are separated
                         by new line, and use 'tab' if the genes are seperate by tabs. Other
                         generic separator are also supported, e.g. ', '. (default: newline)
-  -n , --network        Network to use. {format_choices(config.ALL_NETWORKS)} (default: STRING)
-  -f , --feature        Types of feature to use. The choices are: {Adjacency, Embedding,
-                        Influence} (default: Embedding)
-  -g , --gsc            Geneset collection used to generate negatives and the modelsimilarities.
-                        The choices are: {GO, DisGeNet} (default: GO)
-  -s , --small_edgelist_num_nodes
-                        Number of nodes in the small edgelist. (default: 50)
   -dd , --data_dir      Directory in which the data are stored, if set to None, then use the
                         default data directory ~/.data/geneplexus (default: None)
+  -n , --network        Network to use. The choices are: {BioGRID, STRING, IMP} (default: STRING)
+  -f , --feature        Types of feature to use. The choices are: {SixSpeciesN2V} (default:
+                        SixSpeciesN2V)
+  -s1 , --sp_trn        Species of training data The choices are: {Human, Mouse, Fly, Worm,
+                        Zebrafish, Yeast} (default: Human)
+  -s2 , --sp_res        Species of results data The choices are: {Human, Mouse, Fly, Worm,
+                        Zebrafish, Yeast} (default: Mouse)
+  -g1 , --gsc_trn       Geneset collection used to generate negatives. The choices are: {GO,
+                        Monarch, Mondo, Combined} (default: GO)
+  -g2 , --gsc_res       Geneset collection used for model similarities. The choices are: {GO,
+                        Monarch, Mondo, Combined} (default: GO)
+  -s , --small_edgelist_num_nodes
+                        Number of nodes in the small edgelist. (default: 50)
   -od , --output_dir    Output directory with respect to the repo root directory. (default:
                         result/)
   -l , --log_level      Logging level. The choices are: {CRITICAL, ERROR, WARNING, INFO, DEBUG}
                         (default: INFO)
+  -ad, --auto_download_off
+                        Turns off autodownloader which is on by default. (default: False)
   -q, --quiet           Suppress log messages (same as setting log_level to CRITICAL). (default:
                         False)
   -z, --zip-output      If set, then compress the output directory into a Zip file. (default:
                         False)
   --clear-data          Clear data directory and exit. (default: False)
   --overwrite           Overwrite existing result directory if set. (default: False)
-  --skip-mdl-sim        Skip model similarity computation. This computation is not yet available
-                        when using custom networks due to the lack of pretrained models for
-                        comparison. (default: False)
+  --skip-mdl-sim        Skip model similarity computation (default: False)
+  --skip-sm-edgelist    Skip making small edgelist. (default: False)
 ```
 
 # Dev

docs/source/index.rst

@@ -17,6 +17,7 @@ PyGenePlexus
    notes/api
    notes/r
    notes/data
+   notes/faqs
 
 .. toctree::
    :maxdepth: 1

docs/source/notes/data.rst

@@ -17,6 +17,33 @@ Features :term:`SixSpeciesN2V`
 GSCs     [GO]_, [Monarch]_, [Mondo]_, Combined
 ======== =======================================================
 
+**Detailed species info:**
+
+.. list-table::
+   :widths: 10 10 10
+
+   * -
+     - Specifc Name
+     - Taxon Id
+   * - Human
+     - Homo sapiens
+     - 9606
+   * - Mouse
+     - Mus musculus
+     - 10090
+   * - Fly
+     - Drosophila melanogaster
+     - 7227
+   * - Zebrafish
+     - Danio rerio
+     - 7955
+   * - Worm
+     - Caenorhabditis elegans
+     - 6239
+   * - Yeast
+     - Saccharomyces cerevisiae
+     - 4932
+
 Due to the availability of the data, the following combinations are supported:
 
 .. list-table:: Available Network Options

docs/source/notes/faqs.rst

@@ -0,0 +1,22 @@
+PyGenePlexus FAQs
+=====================
+
+Frequently Asked Questions
+--------------------------
+
+**How are positive and negative genes determined?**
+
+In the supervised machine learning model, any gene from the user-supplied
+gene list that is able to be converted to an Entrez ID and is also in the network is
+considered part of the positive class.
+
+Genes in the negative class based on the chosen Geneset Context. The default Geneset
+Context is Combined, which used all avilable geneset collections.
+
+GenePlexus then automatically selects the genes in the negative class by:
+
+#. Considering the total pool of possible negative genes to be any gene that has an annotation to at least one of the terms in the selected geneset collection.
+#. Retaining all terms in the selected geneset collection that have between 10 and 200 genes annotated to them.
+#. Removing genes that are in the positive class.
+#. Performing a hypergeometric test between the genes in the positive class and the lists of genes annotated to every term in the selected geneset collection. If the value of this hypergeometric test is less than 0.05, all genes from the given term are also removed from the pool of possible negative genes.
+#. Declaring all the remaining genes in the pool of possible negative genes as the negative class.

geneplexus/__init__.py

@@ -2,10 +2,10 @@
 
 .. currentmodule:: geneplexus.GenePlexus
 
-PyGenePlexus enables researchers to predict novel genes similar to their
-genes of interest based on their patterns of connectivity in genome-scale
-molecular interaction networks, and addtionaly translate these findings
-across species.
+PyGenePlexus enables researchers to predict genes similar to an uploaded
+geneset of interest based on patterns of connectivity in genome-scale
+molecular interaction networks, with the ability to translate these
+findings across species.
 
 .. figure:: ../figures/mainfigure.png
   :scale: 20 %