Empowering Modern Biologists with Data Science Skills

Biological data has exploded in volume and complexity, pushing all biologists to become data scientists.

This shift diverts their focus from scientific discovery to data management and analysis. ViaScientific’s platform Foundry is transforming this landscape by handling biological data 20 times more efficiently, allowing scientists to return to their core expertise.

The Data Dilemma

Historically, biologists managed their data in notebooks, manually recording observations and experiments. The advent of computers revolutionized this process, making data tracking and basic computations easier. However, as biological research advanced, the data generated grew exponentially, and traditional data management tools couldn’t keep up.

Today, biologists often spend more time on data processing than on generating new scientific ideas. Complicated, time-consuming and error prone processes hinder reproducibility and progress. When new ideas emerge, researchers often find themselves starting from scratch. Biology turns into a frustrating waiting game.

The Rise of Bioinformatics

To tackle the data overflow, a new profession emerged: bioinformatics. These experts specialize in managing and analyzing biological data using computational tools. Yet even skilled bioinformaticians face the same challenges. The complexity and scale of data can overwhelm manual approaches, leading to inefficiencies and reproducibility gaps.

Via Foundry: A Paradigm Shift

Recognizing the need for change, scientists at UMass Chan Medical School developed Foundry for over a decade — a platform born from their own research needs. This initiative led to the founding of Via Scientific Inc., which today offers Via Foundry as the world’s most advanced multi-omics platform. Via Foundry combines artificial intelligence, user-friendly interfaces, and reproducibility principles to revolutionize data analysis.

Case Study: Reproducibility in AIRR-seq Data Analysis

A recent study¹ published in Briefings in Bioinformatics (Volume 25, Issue 3, May 2024) emphasizes the importance of reproducibility in adaptive immune receptor repertoire sequencing (AIRR-seq) data analysis. The study, authored by Ayelet Peres and colleagues, provides guidelines for achieving reproducibility using Via Foundry.

Reproducibility: The FAIR Pathway

The FAIR principles — Findability, Accessibility, Interoperability, and Reusability — are important for data sharing and analysis. Via Foundry helps researchers follow these principles, making it easier to find, access, share, and reuse analysis pipelines.

• Findability: Clear documentation and standardized metadata help make analysis pipelines easy to find. Via Foundry’s user-friendly interface supports this.
• Accessibility: Easy access to analysis pipelines is crucial. Via Foundry’s versioned containers and thorough documentation make this possible.
• Interoperability: Compatible pipelines allow collaboration across different research groups. Via Foundry ensures pipelines work in various computing environments.
• Reusability: Reusable pipelines are key to reproducibility. Via Foundry’s customizable parameters let researchers adapt workflows to their specific needs, avoiding the need to start from scratch.

Navigating the AIRR-seq Complexity

AIRR-seq data reveals important information about the immune system’s response to infections, diseases, and vaccines. However, analyzing this data is challenging:

• Diverse Production and Tools: Different sequencing platforms and protocols create varied data, requiring many analysis pipelines.
• Non-Standard Reporting: Researchers often leave out important details about their analysis steps, making it hard to reproduce results.

The Via Foundry Solution

Via Foundry addresses these challenges:

• Efficient Data Handling: Via Foundry is 20 times faster than manual methods, freeing scientists from tedious data work.
• Simplified Complex Analyses: Well-documented pipelines ensure consistent, reproducible results.
• Enhanced FAIRness: Via Foundry aligns with the FAIR principles, bridging the gap between data management and scientific discovery.

Simplifying Compliance with Funding Regulations

Via Foundry also helps scientists meet regulations required by funding agencies like the NIH, which mandates a data management plan for all funding applicants. Here’s how Via Foundry supports this:

• Comprehensive Data Management: Via Foundry aids in generating and documenting data types like RNA-Seq, ATAC-Seq, and ChIP-Seq. It ensures data and metadata follow standardized formats, making it easier to share and interpret.
• Integration with Bioinformatics Tools: Via Foundry integrates with popular tools (e.g., FastQC, Bowtie2, STAR) and documents necessary software and versions, ensuring reproducibility through containerized environments.
• Data Preservation and Access: Via Foundry helps researchers choose appropriate repositories (e.g., NCBI GEO, SRA) for data archiving, assign unique identifiers, and set timelines for data availability in line with NIH requirements.
• Legal and Ethical Compliance: Via Foundry manages informed consent, privacy, and confidentiality regulations, controlling access to sensitive data and facilitating data use agreements.

Conclusion: A Brighter Future

With platforms like Via Foundry, biologists can focus on their research without needing to become full-time data scientists. Adopting FAIR principles and prioritizing reproducibility will lead to more groundbreaking discoveries. The future of biology is bright, with efficient, transparent workflows overcoming data challenges.

1. Guidelines for Reproducible Analysis of Adaptive Immune Receptor Repertoire Sequencing Data by Ayelet Peres, Vered Klein, Boaz Frankel, William Lees, Pazit Polak, Mark Meehan, Artur Rocha, João Correia Lopes, Gur Yaari, Briefings in Bioinformatics, Volume 25, Issue 3, May 2024, bbae221