Updates to example-config.yaml, model_generator.py, run_experiment.py, simulator.py, requirements.txt, .gitignore, beeline-inputs-boolean.yaml. Reorganized genVis.py

[mallenjon]

• example-config.yaml: Updated to have proper parameter name recognized by BoolODE for reference. Additional reference comment added.
• model_generator.py: There was a slight issue recognizing the listed proteins/stimuli in the parameters input file. As a result, they would erroneously be given self-activation instead of being treated as parameters. This small syntax change corrects this error so that those proteins/stimuli in the parameters input file can be recognized and treated as parameters as intended.
• run_experiment.py: The n_jobs parameter of sklearn.cluster.KMeans has since been deprecated in version .23 and will later be removed in the new version 1.0 (a renaming of .25) of their k-means clustering package sklearn.cluster.KMeans. This simply removes the use of that parameter from our code. Given this, it would be wise to update requirements.txt accordingly. For more information, see the description for requirements.txt below.
• simulator.py: There was an issue with properly mapping genes with underscores in their titles to the data frame. I have verified that this small syntax change corrects the error so that genes can have underscores in their titles and still be recognized. Verification performed using EMT.txt.
• genVis.py: This is a completely reorganized version of the original genVis.py, the script to dimensionally reduce and visualize BoolODE simulations. Both t-SNE and PCA remain options for dimensional reduction, as well as the added option of UMAP. This updated version allows for simpler arguments from the command line, adds cluster organization, has options for visualization in both 2D and 3D, and allows the user to preview all generated 2D and 3D images as well as move around the latter option for the best picture. The user also has the option to specify a name for the network, which can be any number of words.
• requirements.txt: This is a simple edit to the requirements for installation, which accounts for the event that users may have versions of the dependencies ahead of what is required. The scikit-learn requirement should be greater than or equal to version .23 to account for the deprecated n_jobs parameter in sklearn.cluster.KMeans, and to ensure better performance of our clustering using the new version of sklearn.cluster.KMeans that has accounted for this change. The requirements PyYAML, freetype-py, pypng, and umap-learn are now appended to the file.
• .gitignore: Updated to prevent pushing of any new network data, local config files, local test scripts, and local output files.
• beeline-inputs-boolean.yaml: File renamed from previous typo in name: beeling-inputs-boolean.yaml.

Note: Comments containing requests for edits that have been responded to with a thumbs up 👍 indicate that an attempt has been made to implement the suggestion and/or it is fixed.