Open Research Data Projects

Academic journals have been traditional channels for sharing knowledge. However, the volume of scientific data often surpasses what journals can accommodate. To foster scientific collaboration and discovery, open science advocates the public distribution of scientists' work. To achieve this, we need user-friendly platforms. This project aims to create a platform for the computational solid mechanics (CSM) community. It will allow describing input, code, and simulation output, facilitating storage on a repository. The platform will support mainstream software and high-performance computing calculations. Access and analysis of results will be web-based. When ready for publication, a simple submission to an open-access repository will be the only requirement. While open-source software exists for some aspects of our proposal, CSM-specific needs require a distributed platform, building on initiatives like Meshio and Renku projects.

Recent years have seen a seismic monitoring revolution with fiber optic sensing technologies. Distributed Acoustic Sensing (DAS) utilizes light pulses and backscattered signals to measure strains along fiber optic cables with exceptional spatial precision. The Seismology and Wave Physics group at ETH has pioneered DAS experiments worldwide, covering urban seismic activity, glaciers, avalanches, and volcanic events. DAS generates vast data volumes, necessitating storage, processing, and sharing solutions. As more institutions explore DAS, standards must be established for interdisciplinary collaboration, encompassing FAIR Open Research Data (ORD) and Open Science (OS) principles. This ETH ORD project aims to implement ORD and OS practices across the entire scientific lifespan of geophysical projects, working closely with the community to set an example.

Open data about chemical and material uses are crucial for safety assessments and designing safer alternatives. However, accessing such data in the public domain is challenging due to fragmentation and inconsistent formats. This project seeks to empower the scientific and regulatory communities to create open, FAIR data on chemical and material uses by providing essential tools. The toolbox will include reporting standards and templates, along with cheminformatics and natural language processing workflows for extracting and harmonizing existing data from various sources. These tools are designed for researchers within the ETH-Domain and external stakeholders, including industry professionals, regulators, and civil society organizations. The project will follow a participatory approach, involving ETH-Domain and external stakeholders in testing, finalizing, and sharing the toolbox.

This proposal addresses essential changes in current data treatment and prototyping processes to enhance teamwork, communication, and data visualization for the atmospheric chemistry research community. The motivation stems from the shift toward interdisciplinary collaborations, emphasizing team dynamics, and the growing volume of data. However, effectively exploring and visualizing raw data in the early stages of the FAIR research cycle remains a challenge, which we refer to as the "dark period." During this data acquisition phase, early career researchers and scientists encounter numerous non-standardized data formats and highly diverse raw data from scientific instruments. The objective of this proposal is to investigate FAIR ORD procedures using FAIR services like openBIS (https://openbis.ch/) and Renku (https://renkulab.io/), actively developed by service partners in the ETH-domain. The aim is to prototype procedures for rapid data visualization, enabling collaborative data analysis and discussion among team members, regardless of their expertise.

Scientific publications are valuable, and open research data (ORD) like open access publications have made scientific knowledge more accessible. However, merely accessing this data is insufficient; we must extract and utilize the knowledge within. Traditional keyword-based search engines have limitations. As Natural Language Processing (NLP) techniques advance, we can improve tasks like discovering, reviewing, summarizing, and generating discussions from scientific literature. Our proposal aims to enhance the value of open access scientific content by creating an OR database with over 140 million papers and providing access to cutting-edge NLP tools. We plan to publish the database's code and content, offer an API, and run a web-based application using NLP to assist scientific manuscript authors in assimilating literature. These efforts facilitate knowledge utilization, streamline scientific discovery, and aid in writing processes.

Understanding protein structure is vital for studying protein function mechanisms. The Protein Data Bank (PDB), a prominent repository of protein structures, has been invaluable for 50 years. However, most PDB structures are static, and proteins' functions often involve motion and conformational changes. To achieve a comprehensive view of protein function, we introduced LiP-MS, a structural proteomics tool at ETHZ. It probes dynamic structural changes in proteins in response to various perturbations on a proteome-wide scale. This approach finds applications in biology, biomedicine, and drug development. Our objective is to establish the first open research data infrastructure for storing and sharing this unique dynamic, in situ protein structural data. The infrastructure enhances data accessibility, reusability, and findability. We'll define ORD guidelines for LiP-MS-based dynamic structural data through collaboration (Aim 1), develop an open-access, web-based repository adhering to FAIR principles (Aim 2), and populate it with dynamic structural data from approximately 25,000 proteins across six organisms. This resource will serve diverse user communities effectively.

Modern scientific instruments are intricate systems with hardware, electronics, and software components. To advance the technique, access to all levels – instrument, hardware, software – is necessary. Unfortunately, commercial instrument manufacturers often keep their technology proprietary, limiting access. This hinders innovation, especially in scanning probe microscopy. This project aims to unite the community to share and standardize SPM hardware and software developments. The goal is to establish an open development ecosystem, reducing redundancy in research projects. Existing open hardware AFM projects and early adopters will form the OpenSPM ecosystem, featuring open hardware tools, a knowledge base, global user community, and a sourcing platform for open hardware adoption.

Inadequate data management impedes progress in the Water, Sanitation, and Hygiene (WASH) sector. WASH professionals lack training in essential data management competencies, hindering effective data storage, organization, description, and sharing, in line with FAIR data principles. To address this gap, the openwashdata community will be established to support members. Open-source computational tools will be utilized and taught to enhance FAIR data competencies. Anticipate over 50 previously unreleased data sets contributed by community members. This will foster a network of WASH professionals committed to implementing FAIR data principles, benefiting the entire sector.

Open Research Data (ORD) is fundamental to open science. Unfortunately, publicly available environmental science observations for broader use are scarce, notably in urban drainage research. This is critical because it can lead to significant future cost savings due to high infrastructure expenses and diverse sensor data. The absence of detailed meta-data, such as calibration protocols, complicates data interpretation and fitness assessment. To enhance urban drainage ORD utilization, the researchers propose adapting existing information models like InfraML and WaterML2.0. Standardized dashboards will improve data interpretability and reusability. Engagement with the international urban drainage community will be demonstrated through three European urban drainage datasets. This initiative benefits the ETH domain by fostering synergy with SLF, WSL, and other EXPLORE projects.

This project envisions The Imaging Plaza, an online catalog of FAIR imaging software for ETH domain scientists. The need for enhanced visibility and accessibility of Swiss imaging research output inspires this concept. The Imaging Plaza's goal is to simplify non-experts' access to imaging code produced by peers, enabling confident navigation through available options, providing guidance and incentives. Its aim is not to host code but to facilitate discovery. The joint effort involves the EPFL Center for Imaging and the SDSC, with deployment test sites at EPFL, PSI, and ETHZ. Expert curation ensures software is reusable and well-documented. SDSC engineers create shareable runtime environments. Users can navigate the catalog via a tailored search system and a common language (ontology) for imaging. A "FAIR levels" framework indicates code accessibility for untrained users.