PredictMod Frequently Asked Questions

Go Back to PredictMod Project.

Login & Registration

Who can register for a PredictMod account?

Any clinicians, and researchers, or other individuals interested in using the tool are invited to register.

How long does registration last?

Registered accounts do not expire. You will be logged out after 24 hours of each login.

Will the system store patient data?

The data will not be saved in the system. PredictMod will use uploaded patient data to make a one-time prediction.

Can I view a history of my predictions?

Prediction results are not saved, so a prediction history is not available.

Query Builder

How can I run a prediction using a single-patient data file?

Use the query builder to select the desired condition, intervention, and data type associated with your patient data. File upload templates vary by model. Reference our example files to ensure your data meets our formatting requirements. Select ‘Run Prediction’ to view the results.

What file types are accepted?

Comma separated values (csv) or excel workbook (xlsx) files are accepted.

How do I know if my file is not formatted correctly?

An incorrectly formatted file will return an error message when the 'Run a Prediction' option is selected. 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.

For additional assistance with uploading and formatting data, please contact our team.

What data types are included in the PredictMod query selection?

The current data types are metagenomic (MG), glycomic, glycoproteomic, and electronic medical record (EMR).

• MG data consist of the microbial composition of the gut microbiome typically displayed as a percent of abundance. This data is typically used within a research context to profile healthy gut microbiomes and to identify differential abundances and their associations to disease. Our MG Model is trained on publicly available data. The reference genomes used for this model are recorded in the Gut Feeling Knowledge Base (GFKB). Please use this list as a reference. Organisms not referenced in the GFKB can still be included in the input and receive an accurate model prediction. If you have any questions about this, please don’t hesitate to contact our team.
• EMR consists of real-time patient-centered records that are utilized by physicians to streamline their workflow. It provides physicians the ability to view important medical details about their patients in order to provide improved patient care that is both efficient and safe. Our EMR models are trained on data from the MDClone and Epic Cosmos.

What conditions are included in the PredictMod query selection?

The current conditions of interest are prediabetes, epilepsy, and cancer. The platform is condition-agnostic and is designed to be used for a variety of conditions, interventions, and data types.

• Prediabetes is a precursor to Type 2 Diabetes Mellitus (T2DM), where blood sugars are higher than normal, but not high enough to be considered T2DM. Prediction outcomes are based on a 5% reduction in weight, HOMA-IR, or a diagnosis of T2DM.

What interventions are included in the PredictMod query selection?

The current interventions are exercise, ketogenic diet, and dietary counseling. We also have a Semaglutide intervention coming soon.

What are the possible prediction outcomes?

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.

For additional information, please review the models page on the portal.

Data & Security

Will the system store patient data?

The data will not be saved in the system. PredictMod will use uploaded patient data to make a one-time prediction.