Search for: All records

Recent secure aggregation protocols enable privacy-preserving federated learning for high-dimensional models among thousands or even millions of participants. Due to the scale of these use cases, however, end-to-end empirical evaluation of these protocols is impossible. We present OLYMPIA, a framework for empirical evaluation of secure protocols via simulation. OLYMPIA provides an embedded domain-specific language for defining protocols and a simulation framework for evaluating their performance. We implement several recent secure aggregation protocols using OLYMPIA and perform the first empirical comparison of their end-to-end running times. We release OLYMPIA as open open-source. 

We describe here a flexible protocol for eDNA metabarcoding with Oxford Nanopore's MinIon MK1C platform from sampling to sequencing. The first section summarizes some key steps of sampling and sample preservation for both aquatic and terrestrial environments. The second one describes the DNA extraction protocol with the DNeasy PowerSoil Pro Kit of Qiagen for different types of samples (i.e., soil, liquid and dead plant materials). Our DNA amplification and bead purification protocols are characterized in the third section. Finally, the library prep with the Native amplicon Barcoding Kit 96 V14 (SQK-NBD114.96) and sequencing with the R10.4.1 flow cells (FLO-MIN114) and MinION Mk1C device are presented in the last sections. This protocol has been optimized for protists (microbial eukaryotes) and 18S marker, but should be easily adjustable for other organisms, by modifying the sampling and DNA extraction sections, or other markers, even longer ones such as the full ribosomal operon, if need be. Our goal is to teach/train researchers from different fields and different expertise on obtaining nanopore sequences from environmental samples by guiding them from protocol to protocol, focusing on the key steps, and informing them of the expected results based on our 3 runs performed so far. A comprehensive bioinformatic pipeline to treat the data produced, as well as a methodological article discussing this method and the best ways to use it are in preparation.