How does the Chromeleon Data Agent work?

The Tetra Chromeleon Agent interacts with Chromeleon through Chromeleon 7.2.5 SDK. Internally, it contains three tasks: scanning, generation, and upload.

• Scanning: The Tetra Chromeleon Agent scans the Data Vault selected by the user from Management Console to detect any new or updated injections. If so, the Agent records the Injection URLs to an internal SQLite database.
• Generation: based on Scanning results, the Agent fetches the Injection and its related data, and then serializes the data as JSON files to a local folder.
• Upload: the Agent uploads the JSON files to Tetra Data Lake. After the file is uploaded successfully, the file is removed to save disk space.

Which data items are fetched from Chromeleon?

Tetra Chromeleon Agent extracts the injection and the data associated with it. Here is a brief summary of the data items extracted from Chromeleon.

• Injection
• Sequence
• Instrument Method
• Processing Method
• Signal
• Peak
• Peak Purity
• Chromatogram
• Spectral Field
• Spectrum
• Audit Trail
• Report Template
• Report Variables

Can the Tetra Chromeleon Agent detect injections that are updated?

Yes, the Chromeleon Agent detects the sequences and injections that have been changed since its last scan and upload. The following table lists the fields that the Agent uses to detect changes, and the corresponding action the Agent takes for each update type.

How is the Agent impacted if User Mode is deactivated in Chromeleon?

The User Mode setting in Chromeleon defines whether a username/password authentication is required to access the application (manually or through the SDK). When User Mode is deactivated, username/password/role authentication is not required to connect to and access Chromeleon. However, if any object in Chromeleon (for example, a Data Vault or Folder) has an Access Control List specified and has specific roles assigned to it, then the Agent can't access those objects, and any sequence and injection data in those objects won't be extracted by the Agent.

Why can the user not access a Chromeleon object (e.g. Data Vault or Folder)?

Chromeleon has options to explicitly allow or restrict privileges in individual folders or complete Data Vaults by adding users or access groups to the Access Control List for the data object. However, the privileges of a user are always restricted by the roles assigned to him or her; you cannot extend privileges to a user by allowing roles in the Access Control List. In addition, if a Logon Role is enabled, the privileges of a user or access group are always restricted by the privileges of the specified Logon Role.

If the Agent user/role does not have access control for an object, the Agent will not be able to scan that.

What is the impact of changing the path of a Sequence in Chromeleon?

When a path for Sequence is changed in Chromeleon, if that RAW JSON for the Injections in the Sequence has been previously generated by the Agent, then these Injections will be regenerated and will get a new path in S3. These files will not be a new version of the existing RAW JSONs generated earlier for the same Injections, since the path has changed. These newly generated RAW JSONs will be version 1 in the platform with the new S3 path.

What is the impact of changing the Decode Injection URL setting in Agent UI?

When the "Decode Injection URL" setting is off, then the injection URL will contain URL encoding characters (e.g. space will become %20). When the "Decode Injection URL" setting is on, then the injection URL will not contain URL encoding characters. The URL path is used to create the S3 path and toggling this on/off will result in the creation of a new S3 path depending on if special characters are present. If RAW JSON for an injection is already generated, and the injection changes after altering the "Decode Injection URL" setting, new RAW JSON will be generated in a new S3 path as version 1 of the file (and not a new version of the existing RAW JSON). This may have a downstream impact on protocol execution. Please review the impact before altering this setting. Please reach out to your CSM if you have questions on how this will impact the data in TDP.
An example is when the injection URL in Chromeleon is "chrom://desktop-123/ChromeleonLocal/My Sequence/Run 1/exp12345.seq/100.smp",

• When the "Decode Injection URL" setting is off, the S3 path of RAW JSON will be /desktop-123/ChromeleonLocal/My%20Sequence/Run%201/exp12345.seq
• When the "Decode Injection URL" setting is on, /desktop-123/ChromeleonLocal/My Sequence/Run 1/exp12345.seq

Is the Tetra Chromeleon Agent scalable?

Yes, the Tetra Chromeleon Agent is designed to scale horizontally so that it can handle increasingly larger workloads. By assigning the same Destination ID to multiple Agents, you can configure any number of Agents to upload files to the same location in Amazon Simple Storage Service (Amazon S3). Scalability is limited to the Chromeleon Data Vault level currently, with each Agent able to manage one or more Data Vaults.

For more information, see Set Up the Agent Configuration in the Tetra Chromeleon Agent Installation and Configuration (Version 2.x) guide.

📘

NOTE

An update is planned for a future release that will allow the Agent to detect changes at the sequence level by making it a multiprocess application. This update, which is currently in development and testing, will also provide improved computational power, performance, and throughput.