> For the complete documentation index, see [llms.txt](https://docs.biobox.io/guide/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.biobox.io/guide/how-to/create-delta-reports-to-compare-prioritization-frameworks.md).

# Create delta reports to compare prioritization frameworks

A delta report is a snapshot of your data and prioritization criteria at a specific point in time. By comparing two reports, your team can see exactly what changed in the prioritization and why.

Delta reports help you answer questions like:

* What led to the de-prioritization of a target or a disease-relevant cell population?
* How does a target of interest rank across multiple prioritization frameworks?
* How does my top target hold up when I remove GWAS evidence and focus on internal scRNA and proteomic expression?

And many more.

**Before you start**

Delta reports are built on top of a graph model and its underlying reports, so you'll need both in place first. If you haven't set those up yet, see [configuring graph models](/guide/how-to/create-prioritization-graph-models.md) and [how to create reports](/guide/how-to/create-prioritization-graph-models/generate-reports.md).

**The two types of delta reports**

There are two ways to compare reports, depending on the question you're asking:

1. **Compare archived and live reports in a single model.** Use this to see how rankings shift over time as new data and scientific criteria are added to a model.
2. **Compare one object's ranking across models.** Use this to see how a single target ranks under different scientific criteria and hypotheses, with each model representing a different way of thinking about the problem.

This guide walks through both, using a running example: understanding how the prioritization of gene targets for Amyotrophic Lateral Sclerosis (ALS) has changed, and uncovering how institutional knowledge, prioritization criteria, and data have driven those changes.

***

## Navigating to your delta reports

From the top application toolbar, open the **Graph Models** tab. Then, in the left-hand side panel, select **Delta Reports**.

Click **Create a New Report**. You'll be prompted to choose which type of report you want to build.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F58V5gIKqqiaGKIL0OodX%2Fuploads%2FC4nfJ8QJafgQ5LgFTVHE%2FOpenDeltaReports.mp4?alt=media&token=cbf2870c-54cc-451f-b0c7-d4111f2774dc>" %}

#### Choose your report type

Pick the comparison that matches your question:

* **Comparing Archived vs. live reports** shows how the ranking of objects has changed over time as the data and scientific criteria in that model evolved.
* **One object across multiple models** lets you select several models and see how the different scientific criteria and hypotheses built into each one affect where that object lands.

<figure><img src="/files/0vRizpSwlLkOt6dQl5UL" alt=""><figcaption></figcaption></figure>

The next steps depend on which type you chose. Follow the path that matches your selection.

***

## Comparing archived and live reports in a single model

This report shows how the ranking of objects has changed over time as the data and scientific criteria in that model have evolved.

{% stepper %}
{% step %}

### Step 1: Select Intention

Select "Compare Archived and Live reports in a Single Model"

<figure><img src="/files/J0ZrFmeDIEpKywBzhPfv" alt=""><figcaption></figcaption></figure>
{% endstep %}

{% step %}

### **Step 2: Configure the report**

First, select the **graph model** you want to use for the comparison.

Next, select the **anchor object**. The anchor is the context everything is grounded in. To see how gene targets for a disease have changed over time, you'd set a disease such as Amyotrophic Lateral Sclerosis as the anchor object.

Once an anchor is set, BioBox displays the archived and live reports available for that anchor. Select the reports you want to compare.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F58V5gIKqqiaGKIL0OodX%2Fuploads%2FLQPJfbBKHGCl0lqYuZmZ%2FConfiguringReportComparison.mp4?alt=media&token=2747d9f5-8586-4753-9340-33ae97018a91>" %}
{% endstep %}

{% step %}

### **Step 3: Review the results**

You'll see every ranked object laid out across the reports you selected, which enables you to compare them side by side.

To focus on a single object, select the two reports you want to compare and search for a gene. This shows exactly how that target's ranking changed over time and what contributed to the change.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F58V5gIKqqiaGKIL0OodX%2Fuploads%2FEa7WFtK26GpWmBwROsLq%2FsmallReportComparisonResults.mp4?alt=media&token=31c4a242-9797-4a7a-9efb-dd54b1a3787d>" %}
{% endstep %}
{% endstepper %}

***

## Comparing an object's ranking across models

This report lets you select several models and see how the different scientific criteria and hypotheses built into each one affect where that object ranks.

{% stepper %}
{% step %}

### Step 1: Select Intention

Select "Compare a specific object's ranking across models".

<figure><img src="/files/VSqUpXdp7rEKEqCmPFMo" alt=""><figcaption></figcaption></figure>
{% endstep %}

{% step %}

### **Step 2: Configure the report**

First, select the **object you want to track**, for example a specific gene target you're evaluating for ALS.

Next, select the **models** you want to compare. Each model brings your team's reasoning and scientific criteria, so this is where you decide which frameworks to put side by side. For instance, you might compare a model weighted toward GWAS evidence against one built around internal scRNA and proteomic expression.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F58V5gIKqqiaGKIL0OodX%2Fuploads%2FlpadpQkzttmw0QsM9iSX%2FComparing%20Rankings%20across%20models.mp4?alt=media&token=e90c60cd-bee3-4ac2-823c-7fcf7d94a457>" %}
{% endstep %}

{% step %}

### **Step 3: Review the results**

Search for a target, and you'll see how it ranks in each model, along with the criteria that drove the differences. This makes it clear why a target rises under one framework and falls under another.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F58V5gIKqqiaGKIL0OodX%2Fuploads%2FQBeoEQTXQ2WEW4CEj1Mq%2FextrasmallModel%20Comparison%20Results.mp4?alt=media&token=45fb4c94-e2ff-4484-8dec-ef55436933a3>" %}
{% endstep %}
{% endstepper %}

When you're done, you can save the report for later. All saved reports will be available for exploration on the delta report home page.

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F58V5gIKqqiaGKIL0OodX%2Fuploads%2FsbRyDQVY81UfUTuVVQnh%2FSaving%20Reports.mp4?alt=media&token=3ea084c3-e910-4ba6-9ccd-298688d5bfca>" %}