How Selvita uses KNIME Data Apps to Improve Lab Communication and Speed up Sample Analysis
Read how Selvita uses KNIME data apps for better lab communication and faster turnaround time for sample analyses.
Less than 2 monthsfrom prototype to production
Improved communicationbetween scientists via data apps and automated alerts
Proactive error identification and handlingwithout workflow interruptions
Selvita is a publicly listed research organization that provides drug discovery and drug development services for biotech companies, pharmaceutical companies, and academic partners. With a focus on integrated drug discovery projects, efficient collaboration between chemists and biologists is at the center of every project.
Summary
The challenge:
- Analytical pipelines reliant on manual processes and outdated software.
The solution:
- Developed a custom Laboratory Information Management System (LIMS) with KNIME to manage sample registration, project tracking, operations logging, reporting, and communication.
The results:
- Initial prototype within two weeks, fully operational in less than two months.
- Improved collaboration and clear communication between synthesis chemists and analytical specialists.
Challenge: Communication inefficiencies in Selvita’s analytical pipelines
Selvita has its own analytical pipeline. Chemists in the synthesis labs perform reactions to create molecules that might become potential drug candidates. The output of the reaction is then analyzed by purification specialists. When a sample is sent to purification specialists, they need to record its source, its priority level, and the type of analysis it requires.
Given the size of Selvita with hundreds of chemists and twice as many reactions each day, there was a significant lag in the communication between the two teams.
With their existing commercial software nearing end-of-life, Selvita found themselves at a point where building their own Laboratory Information Management System (LIMS) became the next logical step.
The new system had to smoothly handle registering samples along with key details such as the scientist responsible, the type of analysis required, and molecular structures involved. It also needed to manage uploads of initial analysis results, like PDF files from liquid chromatography, process chemical structures, assign samples to scientists, track communications, log operations, and generate reports.
The team had to think carefully about practical deployment issues, such as how many people would be using the system at once, the expected volume of requests, file-sharing needs, secure storage of sensitive data, and making sure users could access everything easily and securely.
Solution: A custom LIMS using KNIME data apps
The AI and ML team developed a custom LIMS tailored to Selvita's specific requirements. They created an API from KNIME for chemists to raise analysis requests, recognizing that different analysis types require varied inputs. Some need multiple file uploads, while others require molecule descriptions via SMILES or drawing.
Using KNIME components made it easier for the team to dynamically re-execute workflows whenever the input data changed. The system could recalculate expected processing times and request positions in the queue instantly, allowing requesters to see expected queuing times for their analysis.
“We use KNIME components and nested components to reduce redundancy and streamline our workflows.” Fabrizio Ambrogi Senior ML Specialist, Selvita
After finishing the analysis, the team used KNIME to automatically email out key information like sample details, calculated properties, and suggestions generated by AI driven models. Line managers received priority notifications to stay informed immediately. Plus, the entire KNIME workflow was easy to update and kept a clear log of all activities.
Using KNIME to build a frontend UI with data apps helped Selvita simplify request handling significantly. The UI allowed purification specialists to update request statuses, send notes back to requesters, and attach reports upon analysis completion.
The AI and ML team also created a KNIME workflow integrated with a Python script to help detect and manage human errors by validating the input data. If the system finds minor discrepancies, it shows a warning but still allows scientists to proceed. However, if critical issues are detected, the workflow stops completely, clearly notifying users about the specific problem before halting.
Results: Better communication, quicker error resolution, and faster turnaround time for sample analyses
“It would have taken me months to develop something like what KNIME made me do in two weeks as a prototype and two months as a finished product with a UI that people like – with warnings, with looping messages, and everything.” Fabrizio Ambrogi, Senior ML Specialist, Selvita
Developing a custom Laboratory Information Management System (LIMS) with KNIME enabled Selvita to:
- Improve communication between chemists and analytical specialists via automated email alerts, priority approvals, and detailed updates.
- Enable quick error identification and handling, helping chemists address data discrepancies proactively without workflow interruptions.
- Deliver dynamic and tailored user experiences based on analytical requirements, leading to better resource management and faster turnaround times for sample analyses.
- Easily connect processes involving KNIME with other common data technologies, such as Python and Excel.
Learn more about KNIME Business Hub or schedule a call with our life sciences team.