We present a short overview of data-driven tools for inpainting and imputation [M. Buzzicotti et al., 2021, Tianyi Li et al., 2023, Tianyi Li et al., 2023] of turbulent data with emphasis on Uncertainty Quantification and predictability of extreme events, in particular concerning dissipative-based fluctuations for Lagrangian data [Tianyi Li et al., 2023]. The importance of establishing benchmarks and grand challenges for the community, based on a FAIR principles (Findability, Accessibility, Interoperability, and Reuse of digital assets) to foster open-science and open-data approach, will be also shortly discussed, using the example and experience of past and existing turbulent databases.