My movie alignment (MMA)

This tool addresses the beam induced motion (BIM) correction step, which is many times required for data collected with DDDs. MMA is a type of web tool demanding large amounts of computational resources, including I/O. Currently MMA workflow offers Li et al. (2013) and Abrishami et al. (2015) algorithms, which deliver as output a corrected averaged micrograph. Additionally, MMA provides access to all graphical capabilities of Optical Flow 2D (Abrishami et al., 2015), allowing to easily analyze the precise pattern of BIM characteristics of the sample being imaged.


Uploading data

Efficient uploading of very large data sets was a must in the design of MMA. Typically, we may consider the size of a movie to be a few GB, with each project requiring thousands of them. The standard Scipion web upload function using a web browser was modified into a resumable, more robust but not faster upload. A second option is also enabled, allowing any user to run an rsync command, which gets transfer rates over Europe of about 90 MB/s. At this speed, a typical 3TB project takes 10 hours to upload. Currently, we are experimenting with other protocol options to improve our data transfer rate, since we realize that the maximum data bandwidth is often a less limiting factor than the protocol being used for the transfer. Tools like Aspera and Globus Connect are currently under testing. Still, the simplicity of use of rsync, together with its availability in most operating systems, makes it an interesting option to keep.

Data remain at the server for a maximum period of 2 weeks, after which it is deleted.


Embedded help guiding our users through the items to click (steps) to align CryoEM movies with already data and finalized results (2 different datasets).


Starts with importing the movies and then you can run different movie alignment methods, either local or global. A corrected and aligned set of micrographs is generated


Input: movies



Output: Micrographs