The following good practices are recommended for handling large data:

• Visualize data to make it easier to identify potential errors when generating the data. 
• Document the entire data analysis workflow through to quality control and the code used in order to make the path to the published results transparent and ensure reproducibility. With tools such as Docker or Binder, entire computing environments can be documented and shared.
• Automate as many steps as possible to ensure that data meets standards and prevent manual errors.
• Use version control and store raw data in a read-only raw version.
• Describe data with metadata - this helps with subsequent reuse, compilation of datasets and meta-analysis, and is a crucial step in preparing data according to FAIR criteria.