PredictMod User Guide - Revision history

Lorikrammer: /* 'Try it Out' Example Query Builder */

2025-08-28T18:20:37Z

'Try it Out' Example Query Builder

← Older revision		Revision as of 18:20, 28 August 2025
Line 26:		Line 26:
	<p>Example data files for each model are provided in the query builder or on the model page. Once you make your query selection, the platform will indicate which model best fits that selection. </p>		<p>Example data files for each model are provided in the query builder or on the model page. Once you make your query selection, the platform will indicate which model best fits that selection. </p>
	[[File:Image.png\|thumb\|813x813px\|Example Query Builder options to review available models, learn more about the selected model, download an example file, or run an example query.\|none]]		[[File:Image.png\|thumb\|813x813px\|Example Query Builder options to review available models, learn more about the selected model, download an example file, or run an example query.\|none]]
	<p>You can then select 'Review available models' to select other models that match that data type, condition, and intervention. The "Learn more about the model" button takes you to the model page, where you can review information that was captured in the model submitter's BioCompute Object (BCO). The "Download example data file" will prompt you to download a single-patient example file. </p><p>By selecting "Example Analysis", you can run a prediction without having to download the example file.</p>		<p>You can then select 'Review available models' to select other models that match that data type, condition, and intervention. The "Learn more about the model" button takes you to the model page, where you can review information that was captured in the model submitter's [[PredictMod BCOs\|BioCompute Object (BCO)]]. The "Download example data file" will prompt you to download a single-patient example file. </p><p>By selecting "Example Analysis", you can run a prediction without having to download the example file.</p>

	<h2>Current Models</h2>		<h2>Current Models</h2>

Lorikrammer at 18:12, 28 August 2025

2025-08-28T18:12:34Z

← Older revision		Revision as of 18:12, 28 August 2025
Line 7:		Line 7:
	<ul>		<ul>
	<li>[https://hivelab.biochemistry.gwu.edu/predictmod/ PredictMod Portal]</li>		<li>[https://hivelab.biochemistry.gwu.edu/predictmod/ PredictMod Portal]</li>
	<li>[[Frequently Asked Questions]]</li>		<li>[[PredictMod Frequently Asked Questions\|Frequently Asked Questions]]</li>
	<li>[[PredictMod Contact Us\|Contact Us]]</li>		<li>[[PredictMod Contact Us\|Contact Us]]</li>
	</ul>		</ul>

Lorikrammer at 18:11, 28 August 2025

2025-08-28T18:11:34Z

Lorikrammer: /* Automated Pipeline Tool */

2025-08-28T18:10:38Z

Automated Pipeline Tool

← Older revision		Revision as of 18:10, 28 August 2025
Line 57:		Line 57:
	== Automated Pipeline Tool ==		== Automated Pipeline Tool ==
	We have created an [[PredictMod Automated Pipeline\|automated model training pipeline]] for researchers who have a significant collection of data without the programming expertise required to create and train such models. Such researchers may use this automated pipeline to directly upload data.		We have created an [[PredictMod Automated Pipeline\|automated model training pipeline]] for researchers who have a significant collection of data without the programming expertise required to create and train such models. Such researchers may use this automated pipeline to directly upload data.
	[[File:Automated Pipeline Screenshot.png\|none\|thumb\|780x780px]]		[[File:Automated Pipeline Screenshot.png\|none\|thumb\|780x780px\|Automated Pipeline page on the PredictMod platform.]]

Lorikrammer: /* 'Try it Out' Example Query Builder */

2025-08-28T18:10:17Z

'Try it Out' Example Query Builder

← Older revision		Revision as of 18:10, 28 August 2025
Line 13:		Line 13:
	<h2>Login & Registration</h2>		<h2>Login & Registration</h2>
	<h3>How to register with PredictMod</h3>		<h3>How to register with PredictMod</h3>
	<p>Individuals interested in creating a PredictMod account should do so through the [https://hivelab.biochemistry.gwu.edu/predictmod/login Login] page. If you have any questions, please contact us at ''mazumder_lab@gwu.edu''.</p>		<p>Individuals interested in creating a PredictMod account should do so through the [https://hivelab.biochemistry.gwu.edu/predictmod/login Login] page. If you have any questions, please contact us at ''mazumder_lab@gwu.edu''. </p>
			[[File:Login screenshot.png\|thumb\|596x596px\|PredictMod Login and Registration page.]]
			<p></p>

	== 'Try it Out' Example Query Builder ==		== 'Try it Out' Example Query Builder ==
	<h3>How the example query builder works</h3>		<h3>How the example query builder works</h3>
	<p>The example query builder is distinct from the standard query builder in that only the example prediction file is accepted. You cannot upload your own single-patient files into the example query builder and obtain a prediction.</p>		<p>The example query builder is distinct from the standard query builder in that only the example prediction file is accepted. You cannot upload your own single-patient files into the example query builder and obtain a prediction. You can build an example query by selecting the condition, intervention, and data type of interest to you. Based on these selections, an example data file will be available for download.</p>

	~~<p>You can build an example query by selecting the condition, intervention, and data type of interest to you. Based on these selections, an example data file will be available for download.</p>~~		<p>You should use example data files as a template for your own data upload. Data uploads must be in .csv format.</p>

	<p>You should use example data files as a template for your own data upload.</p>

	<h3>Download example data</h3>		<h3>Download example data</h3>
	<p>Example data files for each model are provided in the query builder.</p>		<p>Example data files for each model are provided in the query builder or on the model page. Once you make your query selection, the platform will indicate which model best fits that selection. </p>
			[[File:Image.png\|thumb\|541x541px\|Example Query Builder options to review available models, learn more about the selected model, download an example file, or run an example query.]]
			<p>You can then select 'Review available models' to select other models that match that data type, condition, and intervention. The "Learn more about the model" button takes you to the model page, where you can review information that was captured in the model submitter's BioCompute Object (BCO). The "Download example data file" will prompt you to download a single-patient example file. </p><p>By selecting "Example Analysis", you can run a prediction without having to download the example file.</p>

	<h2>Current Models</h2>		<h2>Current Models</h2>
	<p>Current ~~and anticipated~~ models are shown on the [https://hivelab.biochemistry.gwu.edu/predictmod/models Models] page.</p>		<p>Current models are shown on the [https://hivelab.biochemistry.gwu.edu/predictmod/models Models] page. Click on a model to learn more.</p>

	<h2>Query Builder</h2>		<h2>Query Builder</h2>

	<h3>How the query builder works</h3>		<h3>How the query builder works</h3>
	<p>The query builder determines the appropriate model to use for a prediction, based on the desired condition, intervention, and data type. Please follow the prompts on the Query Builder page to make your selections. Descriptions of the conditions, interventions, and data types are documented within each Model's ~~BioCompute Object (~~BCO). You will then be able to upload your own file or download an example file. The uploaded file must meet the formatting requirements associated with the chosen model. For information on formatting, please review the sample data, FAQs, or contact our team.</p>		<p>The query builder determines the appropriate model to use for a prediction, based on the desired condition, intervention, and data type. Please follow the prompts on the Query Builder page to make your selections. Descriptions of the conditions, interventions, and data types are documented within each Model's BCO. You will then be able to upload your own file or download an example file. The uploaded file must meet the formatting requirements associated with the chosen model. For information on formatting, please review the sample data, FAQs, or contact our team.</p>

	<h3>Run a prediction</h3>		<h3>Run a prediction</h3>
Line 43:		Line 45:

	<h4>Data visualization examples and interpretations</h4>		<h4>Data visualization examples and interpretations</h4>
	<p>The ~~primary~~ data visualization tool is a SHAP force plot. Shapley Value originates from game theory and involves the fair distribution of reward based on the degree of contribution of each player. This can be utilized in precision medicine to identify the key “players” or features that contribute to a given prediction. The SHAP (SHapley Additive exPlanations) Force Plot leverages this ideology to provide Explainable AI with respect to the single-patient predictions made by PredictMod. Each plot for a given prediction not only indicates the most influential features but highlights whether that feature pushes the prediction higher (in red) or lower (in blue). The consideration of the features and their values leads to a score, where higher scores indicate a prediction of 1, or NR and lower scores a prediction of 0, or R. SHAP Force Plots also indicate the degree of feature impact based on proximity to the boundary line where the red and blue bars meet. The closer to the dividing boundary, the more impact that feature had on the patient’s prediction.</p>		<p>The graphs and figures available for each model will vary as it is up to the discretion of the model submitter. The majority of existing models include a confusion matrix and feature importance. You can view these and any other model information on the model page. </p>
			[[File:Model visualization screenshot.png\|thumb\|650x650px\|Confusion Matrix and top feature importance for the ''MDClone Exercise'' model.]]
			<p></p><p>One popular data visualization tool is a SHAP force plot. Shapley Value originates from game theory and involves the fair distribution of reward based on the degree of contribution of each player. This can be utilized in precision medicine to identify the key “players” or features that contribute to a given prediction. The SHAP (SHapley Additive exPlanations) Force Plot leverages this ideology to provide Explainable AI with respect to the single-patient predictions made by PredictMod. Each plot for a given prediction not only indicates the most influential features but highlights whether that feature pushes the prediction higher (in red) or lower (in blue). The consideration of the features and their values leads to a score, where higher scores indicate a prediction of 1, or NR and lower scores a prediction of 0, or R. SHAP Force Plots also indicate the degree of feature impact based on proximity to the boundary line where the red and blue bars meet. The closer to the dividing boundary, the more impact that feature had on the patient’s prediction.</p>

	<h3>Run another prediction</h3>		<h3>Run another prediction</h3>
Line 49:		Line 53:

	<h2>Upload a Model</h2>		<h2>Upload a Model</h2>
	PredictMod is a collaborative space for researchers to upload their intervention-based models and performance metrics. These models are freely available to users and commercial entities under the CC BY 4.0 license. While our current ~~focus is~~ Prediabetes, the platform allows for multiple models to overlap among conditions and interventions. Researchers can upload their model and relevant documentation ~~directly to PredictMod~~ to make it freely available to users.		PredictMod is a collaborative space for researchers to upload their intervention-based models and performance metrics. These models are freely available to users and commercial entities under the CC BY 4.0 license. While our current efforts have been focused on Prediabetes, Epilepsy, and Cancer, the platform allows for multiple models to overlap among conditions and interventions. Researchers can upload their model and relevant documentation to make it freely available to users.

	== Automated Pipeline Tool ==		== Automated Pipeline Tool ==
	We have created an [[PredictMod Automated Pipeline\|automated model training pipeline]] for researchers who have a significant collection of data without the programming expertise required to create and train such models. Such researchers may use this automated pipeline to directly upload data.		We have created an [[PredictMod Automated Pipeline\|automated model training pipeline]] for researchers who have a significant collection of data without the programming expertise required to create and train such models. Such researchers may use this automated pipeline to directly upload data.
			[[File:Automated Pipeline Screenshot.png\|none\|thumb\|780x780px]]

Lorikrammer: /* Automated Pipeline Tool */

2025-04-16T17:47:31Z

Automated Pipeline Tool

← Older revision		Revision as of 17:47, 16 April 2025
Line 52:		Line 52:

	== Automated Pipeline Tool ==		== Automated Pipeline Tool ==
	We have created an automated model training pipeline for researchers who have a significant collection of data without the programming expertise required to create and train such models. Such researchers may use this automated pipeline to directly upload data.		We have created an [[PredictMod Automated Pipeline\|automated model training pipeline]] for researchers who have a significant collection of data without the programming expertise required to create and train such models. Such researchers may use this automated pipeline to directly upload data.

Lorikrammer at 17:44, 16 April 2025

2025-04-16T17:44:24Z

← Older revision		Revision as of 17:44, 16 April 2025
Line 49:		Line 49:

	<h2>Upload a Model</h2>		<h2>Upload a Model</h2>
	~~<p>~~PredictMod is a collaborative space for researchers to upload their intervention-based models and performance metrics. These models are freely available to users and commercial entities under the CC BY 4.0 license. While our current focus is Prediabetes, the platform allows for multiple models to overlap among conditions and interventions. Researchers can upload their model and relevant documentation directly to PredictMod to make it freely available to users.~~</p>~~		PredictMod is a collaborative space for researchers to upload their intervention-based models and performance metrics. These models are freely available to users and commercial entities under the CC BY 4.0 license. While our current focus is Prediabetes, the platform allows for multiple models to overlap among conditions and interventions. Researchers can upload their model and relevant documentation directly to PredictMod to make it freely available to users.

	~~<br>~~		== Automated Pipeline Tool ==
			We have created an automated model training pipeline for researchers who have a significant collection of data without the programming expertise required to create and train such models. Such researchers may use this automated pipeline to directly upload data.

Lorikrammer at 20:00, 12 March 2025

2025-03-12T20:00:45Z

← Older revision		Revision as of 20:00, 12 March 2025
Line 8:		Line 8:
	<li>[https://hivelab.biochemistry.gwu.edu/predictmod/ PredictMod Portal]</li>		<li>[https://hivelab.biochemistry.gwu.edu/predictmod/ PredictMod Portal]</li>
	<li>[[Frequently Asked Questions]]</li>		<li>[[Frequently Asked Questions]]</li>
	<li>[~~https://hivelab.biochemistry.gwu.edu/predictmod/contact~~ Contact Us]</li>		<li>[[PredictMod Contact Us\|Contact Us]]</li>
	</ul>		</ul>

Lorikrammer at 21:14, 5 March 2025

2025-03-05T21:14:51Z

← Older revision		Revision as of 21:14, 5 March 2025
Line 3:		Line 3:
	<h2>What is PredictMod?</h2>		<h2>What is PredictMod?</h2>
	<p>The PredictMod platform utilizes machine learning tools and complex datasets based on electronic health records, gut microbiome, and -omics data to forecast patient outcomes, often in response to treatment for a particular condition. [[PredictMod\|Learn more about the PredictMod Project]]</p>		<p>The PredictMod platform utilizes machine learning tools and complex datasets based on electronic health records, gut microbiome, and -omics data to forecast patient outcomes, often in response to treatment for a particular condition. [[PredictMod\|Learn more about the PredictMod Project]]</p>

	~~<br>~~

	<h4>Additional Resources:</h4>		<h4>Additional Resources:</h4>
	<ul>		<ul>
	<li>PredictMod Portal</li>		<li>[https://hivelab.biochemistry.gwu.edu/predictmod/ PredictMod Portal]</li>
	<li>Frequently Asked Questions</li>		<li>[[Frequently Asked Questions]]</li>
	<li>Contact Us</li>		<li>[https://hivelab.biochemistry.gwu.edu/predictmod/contact Contact Us]</li>
	</ul>		</ul>

	~~<br>~~

	<h2>Login & Registration</h2>		<h2>Login & Registration</h2>
	<h3>How to register with PredictMod</h3>		<h3>How to register with PredictMod</h3>
	<p>Individuals interested in creating a PredictMod account should do so through the Login page. If you have any questions, please contact us at mazumder_lab@gwu.edu.</p>		<p>Individuals interested in creating a PredictMod account should do so through the [https://hivelab.biochemistry.gwu.edu/predictmod/login Login] page. If you have any questions, please contact us at ''mazumder_lab@gwu.edu''.</p>

	~~<br>~~

	~~<h3>Login & registration FAQs</h3>~~

	~~<br>~~

	~~<h4>Who can register for a PredictMod account?</h4>~~
	~~<p>Any clinicians, and researchers, or other individuals interested in using the tool are invited to register.</p>~~

	~~<h4>How long does registration last?</h4>~~
	~~<p>Registered accounts do not expire. You will be logged out after 24 hours of each login.</p>~~

	~~<h4>Will the system store patient data?</h4>~~
	~~<p>The data will not be saved in the system. PredictMod will use uploaded patient data to make a one-time prediction.</p>~~

	~~<h4>Can I view a history of my predictions?</h4>~~
	~~<p>Prediction results are not saved, so a prediction history is not available.</p>~~

	~~<br~~>

	~~<h2>~~'Try it Out' Example Query Builder~~</h2>~~		== 'Try it Out' Example Query Builder ==
	<h3>How the example query builder works</h3>		<h3>How the example query builder works</h3>
	<p>The example query builder is distinct from the standard query builder in that only the example prediction file is accepted. You cannot upload your own single-patient files into the example query builder and obtain a prediction.</p>		<p>The example query builder is distinct from the standard query builder in that only the example prediction file is accepted. You cannot upload your own single-patient files into the example query builder and obtain a prediction.</p>
Line 46:		Line 22:

	<p>You should use example data files as a template for your own data upload.</p>		<p>You should use example data files as a template for your own data upload.</p>

	~~<br>~~

	<h3>Download example data</h3>		<h3>Download example data</h3>
	<p>Example data files for each model are provided in the query builder.</p>		<p>Example data files for each model are provided in the query builder.</p>

	~~<br>~~

	<h2>Current Models</h2>		<h2>Current Models</h2>
	<p>Current and anticipated models are shown on the Models page.</p>		<p>Current and anticipated models are shown on the [https://hivelab.biochemistry.gwu.edu/predictmod/models Models] page.</p>

	~~<br~~>

	<h2>Query Builder</h2>		<h2>Query Builder</h2>

	~~<br>~~

	<h3>How the query builder works</h3>		<h3>How the query builder works</h3>
	<p>The query builder determines the appropriate model to use for a prediction, based on the desired condition, intervention, and data type. Please follow the prompts on the Query Builder page to make your selections. Descriptions of the conditions, interventions, and data types are documented within each Model's BioCompute Object (BCO). You will then be able to upload your own file or download an example file. The uploaded file must meet the formatting requirements associated with the chosen model. For information on formatting, please review the sample data, FAQs, or contact our team.</p>		<p>The query builder determines the appropriate model to use for a prediction, based on the desired condition, intervention, and data type. Please follow the prompts on the Query Builder page to make your selections. Descriptions of the conditions, interventions, and data types are documented within each Model's BioCompute Object (BCO). You will then be able to upload your own file or download an example file. The uploaded file must meet the formatting requirements associated with the chosen model. For information on formatting, please review the sample data, FAQs, or contact our team.</p>

	~~<br>~~

	~~<h3>Formatting FAQs</h3>~~

	~~<br>~~

	~~<h4>What file types are accepted?</h4>~~
	~~<p>Comma separated values (csv) or excel workbook (xlsx) files are accepted.</p>~~

	~~<h4>How do I know if my file is not formatted correctly?</h4>~~
	~~<p>An incorrectly formatted file will return an error message when the 'Run a Prediction' option is selected.</p>~~

	~~<p>If you receive a formatting error message and you are unsure why, please contact our team and we can work with you to resolve the issue.</p>~~

	~~<br>~~

	<h3>Run a prediction</h3>		<h3>Run a prediction</h3>
Line 92:		Line 44:
	<h4>Data visualization examples and interpretations</h4>		<h4>Data visualization examples and interpretations</h4>
	<p>The primary data visualization tool is a SHAP force plot. Shapley Value originates from game theory and involves the fair distribution of reward based on the degree of contribution of each player. This can be utilized in precision medicine to identify the key “players” or features that contribute to a given prediction. The SHAP (SHapley Additive exPlanations) Force Plot leverages this ideology to provide Explainable AI with respect to the single-patient predictions made by PredictMod. Each plot for a given prediction not only indicates the most influential features but highlights whether that feature pushes the prediction higher (in red) or lower (in blue). The consideration of the features and their values leads to a score, where higher scores indicate a prediction of 1, or NR and lower scores a prediction of 0, or R. SHAP Force Plots also indicate the degree of feature impact based on proximity to the boundary line where the red and blue bars meet. The closer to the dividing boundary, the more impact that feature had on the patient’s prediction.</p>		<p>The primary data visualization tool is a SHAP force plot. Shapley Value originates from game theory and involves the fair distribution of reward based on the degree of contribution of each player. This can be utilized in precision medicine to identify the key “players” or features that contribute to a given prediction. The SHAP (SHapley Additive exPlanations) Force Plot leverages this ideology to provide Explainable AI with respect to the single-patient predictions made by PredictMod. Each plot for a given prediction not only indicates the most influential features but highlights whether that feature pushes the prediction higher (in red) or lower (in blue). The consideration of the features and their values leads to a score, where higher scores indicate a prediction of 1, or NR and lower scores a prediction of 0, or R. SHAP Force Plots also indicate the degree of feature impact based on proximity to the boundary line where the red and blue bars meet. The closer to the dividing boundary, the more impact that feature had on the patient’s prediction.</p>

	~~<br>~~

	<h3>Run another prediction</h3>		<h3>Run another prediction</h3>
	<p>Selecting 'Run Another Prediction' will return you to the query builder page, where you can complete steps 1 through 4 to run a new prediction. As a reminder, no patient data is stored in the PredictMod server, so running another prediction will erase any currently displayed results.</p>		<p>Selecting 'Run Another Prediction' will return you to the query builder page, where you can complete steps 1 through 4 to run a new prediction. As a reminder, no patient data is stored in the PredictMod server, so running another prediction will erase any currently displayed results.</p>

	~~<br>~~

	<h2>Upload a Model</h2>		<h2>Upload a Model</h2>

Lorikrammer at 20:57, 5 March 2025

2025-03-05T20:57:37Z

← Older revision	Revision as of 20:57, 5 March 2025
Line 1:	Line 1:
		<small> Go Back to [[PredictMod\|PredictMod Project]]. </small>

		<h2>What is PredictMod?</h2>
		<p>The PredictMod platform utilizes machine learning tools and complex datasets based on electronic health records, gut microbiome, and -omics data to forecast patient outcomes, often in response to treatment for a particular condition. [[PredictMod\|Learn more about the PredictMod Project]]</p>

		<br>

		<h4>Additional Resources:</h4>
		<ul>
		<li>PredictMod Portal</li>
		<li>Frequently Asked Questions</li>
		<li>Contact Us</li>
		</ul>

		<br>

		<h2>Login & Registration</h2>
		<h3>How to register with PredictMod</h3>
		<p>Individuals interested in creating a PredictMod account should do so through the Login page. If you have any questions, please contact us at mazumder_lab@gwu.edu.</p>

		<br>

		<h3>Login & registration FAQs</h3>

		<br>

		<h4>Who can register for a PredictMod account?</h4>
		<p>Any clinicians, and researchers, or other individuals interested in using the tool are invited to register.</p>

		<h4>How long does registration last?</h4>
		<p>Registered accounts do not expire. You will be logged out after 24 hours of each login.</p>

		<h4>Will the system store patient data?</h4>
		<p>The data will not be saved in the system. PredictMod will use uploaded patient data to make a one-time prediction.</p>

		<h4>Can I view a history of my predictions?</h4>
		<p>Prediction results are not saved, so a prediction history is not available.</p>

		<br>

		<h2>'Try it Out' Example Query Builder</h2>
		<h3>How the example query builder works</h3>
		<p>The example query builder is distinct from the standard query builder in that only the example prediction file is accepted. You cannot upload your own single-patient files into the example query builder and obtain a prediction.</p>

		<p>You can build an example query by selecting the condition, intervention, and data type of interest to you. Based on these selections, an example data file will be available for download.</p>

		<p>You should use example data files as a template for your own data upload.</p>

		<br>

		<h3>Download example data</h3>
		<p>Example data files for each model are provided in the query builder.</p>

		<br>

		<h2>Current Models</h2>
		<p>Current and anticipated models are shown on the Models page.</p>

		<br>

		<h2>Query Builder</h2>

		<br>

		<h3>How the query builder works</h3>
		<p>The query builder determines the appropriate model to use for a prediction, based on the desired condition, intervention, and data type. Please follow the prompts on the Query Builder page to make your selections. Descriptions of the conditions, interventions, and data types are documented within each Model's BioCompute Object (BCO). You will then be able to upload your own file or download an example file. The uploaded file must meet the formatting requirements associated with the chosen model. For information on formatting, please review the sample data, FAQs, or contact our team.</p>

		<br>

		<h3>Formatting FAQs</h3>

		<br>

		<h4>What file types are accepted?</h4>
		<p>Comma separated values (csv) or excel workbook (xlsx) files are accepted.</p>

		<h4>How do I know if my file is not formatted correctly?</h4>
		<p>An incorrectly formatted file will return an error message when the 'Run a Prediction' option is selected.</p>

		<p>If you receive a formatting error message and you are unsure why, please contact our team and we can work with you to resolve the issue.</p>

		<br>

		<h3>Run a prediction</h3>
		<p>Once a correctly formatted file has been uploaded, select ‘Run Prediction’ to view your results.</p>

		<h3>Interpreting a prediction result</h3>

		<h4>Responder vs. Non-Responder outcomes</h4>
		<p>PredictMod will provide a prediction categorized as either Responder or Non-Responder. The outcomes associated with the response status vary for each model, though a Responder result is generally associated with a positive health outcome, and the Non-Responder result is generally associated with a negative health outcome.</p>

		<h4>Data visualization examples and interpretations</h4>
		<p>The primary data visualization tool is a SHAP force plot. Shapley Value originates from game theory and involves the fair distribution of reward based on the degree of contribution of each player. This can be utilized in precision medicine to identify the key “players” or features that contribute to a given prediction. The SHAP (SHapley Additive exPlanations) Force Plot leverages this ideology to provide Explainable AI with respect to the single-patient predictions made by PredictMod. Each plot for a given prediction not only indicates the most influential features but highlights whether that feature pushes the prediction higher (in red) or lower (in blue). The consideration of the features and their values leads to a score, where higher scores indicate a prediction of 1, or NR and lower scores a prediction of 0, or R. SHAP Force Plots also indicate the degree of feature impact based on proximity to the boundary line where the red and blue bars meet. The closer to the dividing boundary, the more impact that feature had on the patient’s prediction.</p>

		<br>

		<h3>Run another prediction</h3>
		<p>Selecting 'Run Another Prediction' will return you to the query builder page, where you can complete steps 1 through 4 to run a new prediction. As a reminder, no patient data is stored in the PredictMod server, so running another prediction will erase any currently displayed results.</p>

		<br>

		<h2>Upload a Model</h2>
		<p>PredictMod is a collaborative space for researchers to upload their intervention-based models and performance metrics. These models are freely available to users and commercial entities under the CC BY 4.0 license. While our current focus is Prediabetes, the platform allows for multiple models to overlap among conditions and interventions. Researchers can upload their model and relevant documentation directly to PredictMod to make it freely available to users.</p>

		<br>

← Older revision		Revision as of 18:11, 28 August 2025
Line 14:		Line 14:
	<h3>How to register with PredictMod</h3>		<h3>How to register with PredictMod</h3>
	<p>Individuals interested in creating a PredictMod account should do so through the [https://hivelab.biochemistry.gwu.edu/predictmod/login Login] page. If you have any questions, please contact us at ''mazumder_lab@gwu.edu''. </p>		<p>Individuals interested in creating a PredictMod account should do so through the [https://hivelab.biochemistry.gwu.edu/predictmod/login Login] page. If you have any questions, please contact us at ''mazumder_lab@gwu.edu''. </p>
	[[File:Login screenshot.png\|thumb\|596x596px\|PredictMod Login and Registration page.]]		[[File:Login screenshot.png\|thumb\|596x596px\|PredictMod Login and Registration page.\|none]]
	<p></p>		<p></p>

Line 25:		Line 25:
	<h3>Download example data</h3>		<h3>Download example data</h3>
	<p>Example data files for each model are provided in the query builder or on the model page. Once you make your query selection, the platform will indicate which model best fits that selection. </p>		<p>Example data files for each model are provided in the query builder or on the model page. Once you make your query selection, the platform will indicate which model best fits that selection. </p>
	[[File:Image.png\|thumb\|~~541x541px~~\|Example Query Builder options to review available models, learn more about the selected model, download an example file, or run an example query.]]		[[File:Image.png\|thumb\|813x813px\|Example Query Builder options to review available models, learn more about the selected model, download an example file, or run an example query.\|none]]
	<p>You can then select 'Review available models' to select other models that match that data type, condition, and intervention. The "Learn more about the model" button takes you to the model page, where you can review information that was captured in the model submitter's BioCompute Object (BCO). The "Download example data file" will prompt you to download a single-patient example file. </p><p>By selecting "Example Analysis", you can run a prediction without having to download the example file.</p>		<p>You can then select 'Review available models' to select other models that match that data type, condition, and intervention. The "Learn more about the model" button takes you to the model page, where you can review information that was captured in the model submitter's BioCompute Object (BCO). The "Download example data file" will prompt you to download a single-patient example file. </p><p>By selecting "Example Analysis", you can run a prediction without having to download the example file.</p>

Line 46:		Line 46:
	<h4>Data visualization examples and interpretations</h4>		<h4>Data visualization examples and interpretations</h4>
	<p>The graphs and figures available for each model will vary as it is up to the discretion of the model submitter. The majority of existing models include a confusion matrix and feature importance. You can view these and any other model information on the model page. </p>		<p>The graphs and figures available for each model will vary as it is up to the discretion of the model submitter. The majority of existing models include a confusion matrix and feature importance. You can view these and any other model information on the model page. </p>
	[[File:Model visualization screenshot.png\|thumb\|650x650px\|Confusion Matrix and top feature importance for the ''MDClone Exercise'' model.]]		[[File:Model visualization screenshot.png\|thumb\|650x650px\|Confusion Matrix and top feature importance for the ''MDClone Exercise'' model.\|none]]
	<p></p><p>One popular data visualization tool is a SHAP force plot. Shapley Value originates from game theory and involves the fair distribution of reward based on the degree of contribution of each player. This can be utilized in precision medicine to identify the key “players” or features that contribute to a given prediction. The SHAP (SHapley Additive exPlanations) Force Plot leverages this ideology to provide Explainable AI with respect to the single-patient predictions made by PredictMod. Each plot for a given prediction not only indicates the most influential features but highlights whether that feature pushes the prediction higher (in red) or lower (in blue). The consideration of the features and their values leads to a score, where higher scores indicate a prediction of 1, or NR and lower scores a prediction of 0, or R. SHAP Force Plots also indicate the degree of feature impact based on proximity to the boundary line where the red and blue bars meet. The closer to the dividing boundary, the more impact that feature had on the patient’s prediction.</p>		<p></p><p>One popular data visualization tool is a SHAP force plot. Shapley Value originates from game theory and involves the fair distribution of reward based on the degree of contribution of each player. This can be utilized in precision medicine to identify the key “players” or features that contribute to a given prediction. The SHAP (SHapley Additive exPlanations) Force Plot leverages this ideology to provide Explainable AI with respect to the single-patient predictions made by PredictMod. Each plot for a given prediction not only indicates the most influential features but highlights whether that feature pushes the prediction higher (in red) or lower (in blue). The consideration of the features and their values leads to a score, where higher scores indicate a prediction of 1, or NR and lower scores a prediction of 0, or R. SHAP Force Plots also indicate the degree of feature impact based on proximity to the boundary line where the red and blue bars meet. The closer to the dividing boundary, the more impact that feature had on the patient’s prediction.</p>