FAIRlyz is a scientific data registry with data quality control (QC) and data curation functionality. Below is a list of FAIRlyz frequently asked questions.
Who should use FAIRlyz and for what purpose?
- Scientists use FAIRlyz for various research tasks. These include planning and executing studies, promoting their work, finding funding, collaborating with others, curating data, running quality control (QC) on their data, and sharing data.
- Research institutions utilize FAIRlyz to ensure adherence to data quality and data sharing protocols, thus preserving the integrity and reputability of their research endeavors.
- Researchers and institutions working with machine learning and AI require access to quality datasets for training and use FAIRlyz to evaluate the quality of the data.
You may consult the FAIRlyz About Us page for more information.
Does an investigator have to share their sensitive or restricted-access data?
No. FAIRlyz’ data visitation capabilities allow researchers to analyze sensitive data without sharing it. FAIRlyz is used by researchers who already possess and store sensitive data locally. It does not facilitate data downloads, it does not move sensitive data, it visits the sensitive data. Metadata about the study which is not sensitive data is stored and shared in the central registry to inform others about the quality, provenance, and availability of data.
Does FAIRlyz meet NIST 800-171 security controls?
FAIRLYZ-QC meets the applicable NIST 800-171 security controls, designed for the protection of Controlled Unclassified Information (CUI). However, certain hardware or environmental controls are the responsibility of the administrators of the end-user workstations or compute environment. Information stored in FAIRLYZ.com central registry platforms consists of metadata from public studies or metadata entered by users about their own studies which is intended for public dissemination and is not CUI.
What kind of files and scientific data are supported?
- Tabular data (CSV format) with a Data Dictionary File in dbGaP format. A Data Dictionary Generator is provided. Such data can be clinical, demographic, and phenotype tabular data.
- (Meta)Genomics MultiQC.html report for data from technologies that generate DNA/RNA sequence data. Raw omics data is not processed by QC.
What kind of files and scientific data are not currently supported?
Proteomics or metabolomics data and omics that do not analyze DNA/RNA sequence data are not supported for QC in the current version.
Does FAIRlyz-QC process raw omics data?
No, FAIRlyz-QC does not directly process raw omics data. It requires that the researcher generate and provide a MultiQC.html report generated from raw sequencing data. If no MultiQC file is available the researcher can install multiqc and generate the file with ‘multiqc .’ in the command line in the directory with the FastQC files. It also works with fastp files.
Does FAIRlyz-QC integrate directly with Box, Dropbox, or Google Drive?
No, FAIRlyz-QC does not directly integrate with the Box, Dropbox, or the Google Drive website or its cloud storage. However, if you sync or download your Box, Dropbox, or Google Drive data to your local machine, FAIRlyz-QC can access and process that data from your local drive.
How does FAIRlyz-QC compare to other QC tools like CEDAR or Dryad?
- CEDAR helps researchers prevent metadata errors through the use of standardized templates. Although FAIRlyz supports templates, it can also evaluate data without them. FAIRlyz uses AI and ontology mapping to assess completeness and identify missing information in existing datasets.
- Dryad’s is a repository using CEDAR metadata templates to get data ready for publication. By using Data Visitation, FAIRlyz enables researchers to work with sensitive data locally without needing to share it, facilitating data curation and QC. FAIRlyz focuses on the crucial steps of data curation and QC that happen before data is formally deposited in a repository like Dryad.
Can anyone register a study and its data?
No, FAIRlyz verifies that the researcher is associated with an ROR.org registered institution. When registering a study or its data, the researcher has to confirm he or she is authorized to register it.
How is FAIRlyz registry different to a data catalog?
Data or metadata in data catalogs are usually added by data managers, data curators, librarians, or software applications, not the researcher who owns the data, e.g. https://www.datasetcatalog.nlm.nih.gov/. In FAIRlyz, it is either the researcher who generated the data who registers it, or it is the researcher who reuses published data who registers the curated and reused data, but authorized data managers may also perform this task. Moreover, FAIRlyz, supports data curation and evaluation moving beyond the common features of a data catalog.
How does FAIRlyz use AI?
AI plays a key role in both the study annotation process and QC ontology mapping during data visitation.