Blog

Introducing Code Ocean Agents: Trusted Science for Humans and AI

Alicia Liu — Sat, 04 Oct 2025 11:59:46 GMT

We’re launching Trusted Agents in Code Ocean 4.0, becoming the first computational science platform that allows AI to execute real analyses in a secure environment where every action is versioned, traceable, and compliant, and each result is fully reproducible. The outcome is accelerated science, reproducible results, full provenance, and enterprise-grade governance.

Introducing Code Ocean Models: a unified environment for ML in CompBio

will.patrick@codeocean.com (Will Patrick) — Tue, 10 Sep 2024 04:15:00 GMT

We’re launching new machine learning tools in Code Ocean 3.0, the latest version of our computational science platform. They make it easier for computational biologists, bioinformaticians, and ML engineers to work together in training, validating, and using ML models alongside their existing workflows. They also come with full traceability and lineage from development to pre-production.

Pre-production: the missing ML link in Biotech & Pharma

Daniel Koster, PhD — Tue, 10 Sep 2024 04:01:00 GMT

There’s a huge gap between ML model hype and adoption because developers can’t—or don’t—involve domain experts early enough.

Map of foundational models for use in biotech and pharma R&D

Daniel Koster, PhD — Wed, 26 Jun 2024 16:00:00 GMT

New foundational models are published all the time. Knowing where to start can be daunting. It can also be an overwhelming task to keep track of new models as they're released, let alone find the time to even use them in an already busy schedule.

Code Ocean and Allen Institute announce partnership - Code Ocean

Code Ocean — Tue, 14 Nov 2023 05:00:00 GMT

Today we officially announced our ongoing partnership with the Allen Institute, a Seattle-based independent nonprofit bioscience and medical research institute founded by Microsoft co-founder and philanthropist Paul G. Allen. Composed of five scientific units and more than 800 employees, the Allen Institute conducts large-scale research through foundational science to fuel the discovery and acceleration of new treatments and cures for diseases such as Alzheimer’s disease, heart disease, cancer, addiction, and more.

In keeping with its belief in open science, one of the Allen Institute’s unique core values is to make all data and resources publicly available for external researchers and institutions to access and use. Code Ocean shares this deep commitment to scientific discovery and open collaboration, and it has served as the foundation of our partnership.

The Catch-22 of Computational Research

Research organizations such as the Allen Institute often face a dilemma when it comes to managing large-scale computational research and computational analysis.

Most computational platforms cater to technical users who are well-versed in a command-line interface, git, docker, APIs, and version control. Research organizations, on the other hand, typically employ scientists who can write experimental code but may not possess deep expertise in engineering. And therein lies the paradox for research organizations — the people handling massive amounts of data are not necessarily equipped to manage the very complex technical aspects associated with generating insights from their datasets. Meanwhile, the individuals who do have this training (engineers, developers, administrators) aren’t necessarily familiar with the research topics and processes.

Historically, this paradox has required research organizations to deploy teams of research scientists and engineers to work in parallel. Although there may be some short-term advantages to this model, most organizations find that it ultimately leads to an inefficient allocation of time and resources.

Ultimately, for scientists to do efficient, collaborative, and fully reproducible computational research while focusing on the science, research organizations need to implement a computational platform that automates the technical complexity involved with large-scale scientific computing. In addition to affording optimal flexibility, taking this step can enable organizations to reallocate engineering resources to more strategic activities.

A Match Made in Data Heaven

When it comes to implementing a platform for computational research, it’s crucial to select the right partner. Naturally, this requires aligning the proficiency of your staff with the platform’s technical capabilities, but it’s equally important to make sure that your technical partner understands and embraces your research organization’s larger objectives. The Allen Institute found the right match in Code Ocean.

Since implementing Code Ocean one year ago, the Allen Institute has achieved dramatic improvements in reproducibility, interoperability, and productivity. The organization has been able to complete more with less and at a quicker pace, enabling more than 100 researchers to actively use the platform to make discoveries faster and more efficiently.

Using Code Ocean has resulted in efficiency gains as well as a reallocation of time and resources to higher-leverage activities:

The speed at which a pipeline can be built has decreased from 12 to 3 weeks, representing— a 4x increase in workflow efficiency.
The Allen Institute has been able to reallocate the time of 5 engineers to more impactful projects.
With fully interoperable Code Ocean Pipelines, the time and effort required to share a pipeline has been reduced to just one click.

Code Ocean’s “no lock-in” platform has also helped the Allen Institute stay true to its mission of advancing open science by reaching and collaborating with more than 200 external users.

Helping Research Organizations to Scale Effectively

Code Ocean has developed a wide range of dedicated features to assist large organizations such as the Allen Institute to operate effectively at scale. These features include fine-grained custom metadata management and search, data organization in Collections, automated data provenance for trusted science and reproducibility, and integration with external data warehouse providers and leading machine learning platforms. One noteworthy statistic is the number of times scientists committed code to git (2,370 times) and the percentage of users committing to git (61%). These figures surpass what’s typically observed within the broader scientific community (Ram, 2013). This achievement has been made possible by Code Ocean’s dedicated UI features, which bypass the need for command-line operations requiring expertise in git.

These statistics show that the Allen Institute not only reaps the benefits of implementing these foundational technologies at a broad scale but also reinforces industry-standard best practices. Among other benefits, Code Ocean has empowered the Allen Institute to utilize machine learning on a broad scale, for the first time handling petabyte-scale data, with a growing user base of 282 users running more than 15,000 computations, totaling over 10,000 hours.

“Our partnership with Allen Institute has helped us to refine our ability to accommodate the specialized needs of large organizations,” said Dr. Daniel Koster, Vice President of Product at Code Ocean. “We have adapted our services to meet requirements such as handling petabyte-scale data, deploying FAIR-based analysis for reliable, scalable science, ensuring the reproducibility of scientific outcomes from experiment initiation to final presentation, complying with necessary regulatory standards, and providing a single pane of glass that integrates the multitude of tools utilized internally.”

Less than a year into their partnership, the Allen Institute has already generated a data corpus on Code Ocean of over a petabyte, with a growing user base of 282 running over 15,000 computations totaling over 10,000 hours.

For a more extensive report on how Code Ocean is accelerating neuroscience research for the Allen Institute, read the full case study.

To learn more about how Code Ocean can help your team, schedule a demo with us!

Lineage Graph: see a visual history of all your computational results

Code Ocean — Thu, 13 Jul 2023 04:00:00 GMT

In the world of scientific research, reproducibility has long been a persistent challenge. A survey conducted by Nature found that more than 70% of researchers encountered difficulties when attempting to reproduce experiments conducted by their peers, and more than half have struggled to reproduce their own work. This issue hampers team progress and collaboration, and for decision makers it introduces doubt into the scientific process.

Code Ocean, now with a new Result Lineage feature, delivers trusted science by offering a visual representation of the complete history of every data result generated within its Digital Lab. This first-of-its-kind feature addresses the disconnect from original source data that plagues computational research. As research data travels from scientist to scientist, it becomes scattered across various tools and locations, making it difficult to share data and code — and almost impossible to reproduce results.

Simon Adar, Co-founder and CEO of Code Ocean, emphasizes the significance of Result Lineage as a solution to the ‘last mile’ problem in computational research. “Scientists spend too much of their time and energy struggling to reproduce previous results to determine whether they are valid. Result Lineage guarantees the traceability and reproducibility of scientific results by illustrating the entire history of each scientific result at a glance. This enhances both the speed and quality of scientific discoveries. It means results presented to decision makers can be trusted, as every result is generated with full lineage for traceability and reproducibility.”

Every data asset generated in Code Ocean’s platform has a result provenance metadata box containing information about how the result was created, where the input data came from, and how it has been used. This metadata is automatically generated. Result Lineage takes this information and presents it in a color-coded graph, so researchers can understand the reproducibility of a result at a glance. Each point on the lineage graph links back to the original source data, providing an easy way to trace its origin and track its usage history. Result Lineage currently covers the entire backward provenance of a result, including the input data, Capsules and pipeline output. Any computational step in the Result Lineage can be reproduced with the click of a button, making the process highly streamlined and efficient.

Code Ocean Lineage Graph showing Result Provenance of a Bulk RNA-Seq Pipeline. A visual representation of the input Data, Capsules, and Pipeline used to generate the final result, “Star Output GSE157194”. Each node of the lineage graph includes a direct link to the asset used for analysis.

David Feng, Director of Scientific Computing at the Allen Institute for Neural Dynamics adds, “We often talk about the importance of provenance tracking, but Code Ocean makes it a transparent part of the analytical workflow. The lineage graph gives us confidence that we will always be able to reproduce our work.”

By addressing the long-standing challenges of traceability, reproducibility, and reuse, Code Ocean continues to empower researchers to accelerate their discoveries, collaborate effectively, and ensure the trustworthiness of their work.

See Result Lineage in action for yourself. Request a demo today.

Code Ocean Wins Gold at 18th Annual Globee® Awards - Code Ocean

Code Ocean — Wed, 31 May 2023 04:00:00 GMT

Code Ocean recently took home the prestigious Gold Award in the 18th Annual 2023 Globee® Awards for Information Technology, securing the title of “Best IT Solution for Pharmaceuticals and Biotech”.

At Code Ocean, we want nothing to slow science down. That’s why we created our digital lab for faster, trusted computational science — bringing together decision-makers, research IT, coding, and domain scientists in one shared, collaborative space — to make science move faster.

“Code Ocean organizes the creative chaos inherent in computational science,” said Simon Adar, Co-founder and CEO, Code Ocean. “Scientists use many well developed methods for capturing, storing, and indexing primary research data, but once that data leaves the security of a database to be transformed into valuable insights, it often loses the connection to its original source. Code Ocean’s digital lab guarantees the traceability and reproducibility of any result so scientific discoveries can be completely trusted by decision makers.”

The Globee Awards honors organizations of all types and sizes from around the world for their achievements in various business and technology-related categories. With a panel of 600+ industry experts from around the globe, nominees for the 2023 Globee® Awards for Information Technology go through a rigorous evaluation process. Code Ocean’s impact on the pharmaceutical and biotech industries stood out, earning us the gold award. It’s an incredible honor!

“The pharmaceutical and biotech industries are going through digital transformations to modernize infrastructure and enable scientists to better collaborate,” said Ram Dayan, Co-founder and Chief Technology Officer, Code Ocean. “Code Ocean is a critical part of that transformation, as it will make more information available for analysis and improve the quality and speed of science. By offering a unified platform for data and computational research in the wet and dry lab, Code Ocean provides full knowledge of the path of discovery, making each stage of the computation fully reproducible.”

As these industries continue to evolve, Code Ocean remains at the forefront, delivering the first and only integrated science library and workbench that guarantees reproducibility and speeds up discovery. We are opening doors to endless possibilities for faster, groundbreaking discoveries.

Request a demo to see how Code Ocean can help accelerate your pharmaceutical and biotech research.

Code Ocean Launches New Apps Library | Code Ocean

Code Ocean — Tue, 25 Apr 2023 04:00:00 GMT

Today, Code Ocean launched a new Apps Library, which offers a variety of customizable Open Source applications. This new toolkit provides code-free access to pre-vetted Code Ocean Capsules — a suite of ready-to-use tools and best-in-class customer support. With 30 pre-vetted Capsules available at launch (and more being added), it provides researchers with a popular selection of customizable biotech and life sciences applications.

In the Apps Library you will find curated, off-the-shelf applications that speed up your science — all reviewed and verified by the Code Ocean Science Teams. Code Ocean’s customers will also have the ability to use pipelines, analysis, and visualizations with a single-click in a secure environment on their private data, and make changes and customizations to their own needs.

“Code Ocean is dedicated to providing new ways for scientists to jumpstart research without having to duplicate steps on IT setup or coding,” said Simon Adar, Co-founder and CEO, Code Ocean. “We’ve identified the most desirable analysis types across our customer base and decided to create a curated ready-to-go extensive collection in bioinformatics. With our new Apps Library, Code Ocean is offering some of the most useful open source tools as out-of-the-box applications. Using these research capsules, which have been carefully reviewed and verified by the Code Ocean’s Science Support Teams, will significantly reduce the time and effort required to perform analyses. One of the unique properties of the Apps Library is that scientists can modify and customize this collection to their specific needs.”

The Apps Library will contain applications relating to many fields of biotechnology, including Alphafold, ShinyCell, DECIMER 2.0, BWA, Star, and more.

“All of these applications can run seamlessly on the cloud from a user-friendly interface which exposes common parameters and inputs,” said June Snedecor, Head of Application Science, Code Ocean. “The Code Ocean Apps Library will be continuously maintained and updated with new applications and topics over the course of coming weeks and months.”

For a full list of current applications, visit the Code Ocean Apps Library. To see the Apps Library in action, watch our webinar, Breaking Barriers: Speed Up Time To Your Next Scientific Breakthrough, available on-demand.

Code Ocean and Nature Portfolio Join Forces - Code Ocean

Code Ocean — Fri, 17 Feb 2023 05:00:00 GMT

Today, Code Ocean announced some exciting news. We’ve teamed up with Nature Portfolio to launch the Nature Portfolio-Code Ocean Open Science Library.

This library contains a curated collection of research software published by authors in popular Nature journals, including Nature Methods, Nature Biotechnology, Nature Computational Science, and Nature Machine Intelligence. And the best part? You can access the library for free by setting up an account at nature.codeocean.com

Life science researchers are all too familiar with the rapidly growing amount of data and computational software that is not immediately reproducible and usable for future research. It’s no surprise that more than 70% of computational research is non-reproducible (according to an analysis by Proceedings of the National Academy of Sciences). That’s where Code Ocean comes in.

Code Ocean makes it easier for researchers to collaborate and work more efficiently through the use of Compute Capsules. These Capsules create a software package with code, data, and computing environment, and automatically trace the path of research data. This makes each stage of computation fully reproducible and gives scientists a complete understanding of their discovery journey.

Nature Portfolio serves the research community by publishing its most significant discoveries, including primary research, reviews, critical comment, news, and analysis. Erika Pastrana, Editorial Director, Nature Portfolio, said, “We’re thrilled to partner with Code Ocean. This unique collection of journal Compute Capsules will enable readers to easily re-use code associated with our publications, and will support our authors to gain more visibility, boost the usage of their work, and ensure reproducibility.

Simon Adar, Co-founder and CEO of Code Ocean, agrees, “Code Ocean addresses the creative chaos inherent in computational science by providing one end-to-end platform for data and computational research in the wet and dry lab. Nature’s highly esteemed research authors publish the world’s cutting-edge computational research on the Open Science Library. This is open access to all researchers. With a library of ready-to-run research software, our joint offering will help to accelerate the speed and enhance the quality of life science research.”

Benjamin Haibe-Kains, Ph.D., Associate Professor of the Medical Biophysics Department at the University of Toronto, added: “If you have to recreate from scratch the analysis code or if you have issues running the provided code, it will likely be a labor-intensive, lengthy, and risky initiative. Code Ocean totally solves tracking and reproducing analysts for researchers and increases trust in the results.”

You can see and use the Nature-Code Ocean Open Science Library for yourself at https://nature.codeocean.com/

5 Reasons to Bring Web Apps Into Your Bioinformatics Work

Code Ocean — Tue, 06 Dec 2022 05:00:00 GMT

Many research teams share static results between working group members, like image files or documents, but this can slow progress down. Feedback can be difficult to give and difficult to incorporate. Next time a bioinformatics project is wrapping up, think about dynamic result sharing with a web app! Here’s why:

No more emailing documents back and forth! Bioinformaticians have to go back to code repeatedly over a project’s lifecycle. They need feedback from team members who have varying technical skills, so usually after they code everything needed for a figure or report, discussion moves off their working platform and into email attachments. This can slow everyone down — colleagues have to manually review graphs, and bioinformaticians have to comb back through their code to incorporate changes. Web apps allow colleagues, at all technical levels, to interact with the data via drop down menus, text boxes, buttons, and sliders.
Web apps encourage proactive experimental planning. Deeply considering the analytical framework for one task will make it easier to prepare for the next. Even using user-friendly tool sets, apps can be an investment — they should be made carefully and align with future work or incoming data. This is a two-way street and gives structured foresight for future data ingestion and even experimental planning: having discussions in the planning phase between all parties that are going to touch the data ultimately makes your science better.
You can standardize plot elements. Even among highly experienced bioinformaticians, people really vary in how they set up axes, select color schemes, mark error bars, etc. This variation can cause confusion, especially for colleagues who need to make quick, clear inferences from these plots. It’s hard to establish standards on these details without extremely lengthy documentation. Building a web app, with core plots and interactive elements, provides a good framework to make standards actionable.
Design is fun and encourages empathy. Design meetings are where you get to bust out the white board! It gets everyone involved on one common task, using visual, symbolic language. Designing an app together involves discussing and understanding how users will look at, and move through, the visual cues in front of them and how they’ll want to interact with the framework. Too often, people can get siloed and frustrated at each other instead of at the structural framework that they’re working with.
There’s a lot of neat tools to choose from! If you’re a data-centric R team: Shiny. Python? Check out Dash by Plotly, Viola, or Streamlit. There are many others for a huge variety of languages and requirements. What they all have in common is active communities — if you have questions or develop work you’re excited to share, there are many dedicated forums where you can get support and cheers. Whatever your team’s existing tools and preferences are, we’re confident you’ll find a web app option that fits.

If you enjoyed reading about web apps and would like to learn more, check out the on-demand webinar, How to Build, Run, and Deploy Bioinformatics Apps to the Cloud in Minutes. Code Ocean bioinformatician Emily So illustrates an end-to-end web app process using Shiny on the Code Ocean platform!