RMaP Nanopore Challenge
RMaP and associated Researchers acknowledge the growing importance of nanopore-based sequencing in RNA science. In particular, the identification of RNA modifications in single molecules is of current interest in the community. A distant goal shared by many is the deciphering of several different modifications in long RNA reads, which is commonly acknowledged as a door opener to a new and rewarding facete of epitranscriptomics.
We are aware of many colleagues developing similar or entirely idiosyncratic approaches to the common problem of enlarging the chemical repertoire of pentanucleotide permutations that crucially define signal shape and information readout in this technology. Once achieved on the molecular level, correct base-and-modification calling will require new bioinformatic concepts and tools. Since the community cannot wait for commercial developments, joint and directed, yet competitive, efforts offer a promising vector along which to move to the common goals defined above.
The RMaP consortium is poised to benefit from such developments and therefore wants to promote them. We therefore decided to launch a challenge, whose first issue aims to identify the most successful data treatment approach of datasets provided for download on this website.
While we have received positive feedback from a required minimum of groups working in this field, we would like to include as many colleagues as reasonably possible, providing a platform for discussion among us and potentially with relevant enterprises. The exact format of the challenge is therefore NOT YET fully DEFINED, in order to give you, as participants, the possibility to shape it.
The rough outlines are as follows:
Datasets and corresponding tasks may be contributed by registered participants, provided they meet a set of plausibility criteria. Data and task should be uploaded onto this website, and will be made available to registered participants, who should submit a solution within a specified time frame. The submitted solutions will be reviewed and ranked by the contributing labs, who can obviously not be competititors in their own task, but may participate in tasks contributed by other labs.
We are currently planning a minimum of two tasks for which data will be available according to the following (preliminary) timeline:
Registration Deadline: June 30th 2023 -Closed-
Task upload: July 31st 2023
Challenge data download open: August 31st 2023
Solution upload closes: October 31st 2023
Scores published: End of 2023
Joint publication: in 2024
We will also need to create some form of scientific board for joint decision making. Details to be announced.
For completeness, you find a listing of both challenges available below:
(1) Gerber/Helm Lab Challenges on Gm
https://cloud.trr319-rmap.de/index.php/s/2wffFCmbwoEmEZQ
(2) Marz Lab Challenges on different modifications
https://cloud.trr319-rmap.de/index.php/s/W6WmLdD7jzaWGpD
For the latter some new data was added mid-september. In case you have not done, please go back and check.
For general information on how you should organize your data for submission,
please visit: https://cloud.trr319-rmap.de/index.php/s/jZar6RXZ9jHPEBt
We have also adjusted the submission deadlines for your convenience.
Submission opens: October 1st 2023
Submission closes: October 31st 2023 (was September 30th before).
Upload and instructions (October 1st 2023-October 31st 2023)
– Login to https://cloud.trr319-rmap.de with your login credentials
– For each challenge create a new folder in your cloud home directory by clicking on the “+” button (choose new folder).
– For example, if you participated in both challenges you create to 2 new folders called: RMaP_challenge_GerberHelm_DataSet and RMaP_challenge_ManjaMerz_DataSet
– You upload your submission to the respective folders, check for completeness and share the respective folders with a public link only.
– Copy this link, keep it safe and send it in an email to christoph.dieterich@uni-heidelberg.de – with Sbjct line: RMaP submission completed
Please consult a Nextcloud manual if in doubt.
