Campuses

As France's leading biocluster, Genopole is an incubator for cutting-edge projects in biotechnology. Located in the city of Évry, just south of Paris, Genopole provides a unique environment for scientists and entrepreneurs seeking to advance research and innovation.

Discover >

Advantages

Genopole accompanies researchers, postdocs and start-up entrepreneurs through all the phases of their projects to ensure the best possible conditions for business development.

Discover >

Genopole’s citizens

Every day, at Genopole, researchers, entrepreneurs and students cross paths, share ideas and unite forces in a veritable melting pot for innovation.

Discover >

Highlights

Giving wings to research and empowering employment in our community are cornerstones of Genopole's mission. Catch up on recent scientific advances, the accomplishments of our biotech actors and the events that enliven the biocluster.

Discover >

Innovate with us

Discover >
Success stories

Plateform EvryRNA, a wide range of tools to accelerate non-coding RNA discoveries


Plateform EvryRNA Plateform EvryRNA

EvryRNA is a Genopole computer program platform that provides analytical tools aimed at the study of non-coding ribonucleic acids (ncRNAs). Research done over the last few decades has revealed the major roles played by ncRNAs not only in biological mechanisms but also in diverse diseases including certain cancers. Under the direction of Prof. Fariza Tahi, the EvryRNA platform is a part of the IBISC laboratory (Computer Science, Bio-informatics and Complex Systems – University of Évry-Paris Saclay). EvryRNA is one of Genopole’s shared-use technological and ancillary services platforms.

Fariza Tahi’s bioinformatics team develops algorithms and computational methodologies specifically for RNA analyses. The team’s aim is to identify RNAs in genomic sequences, determine their coding or non-coding nature and predict their spatial conformation, a key aspect of RNA function. Indeed, RNA is able to fold onto and about itself thanks to the formation of hydrogen bonds between its constituent bases.

Non-coding RNA: a large field of scientific research

The last twenty years have seen the discovery of a multitude of ncRNAs that arise from genome regions long thought to be inactive because they do not code for proteins. Biologists have had a firm grasp on the roles of the non-coding ribosomal and transfer RNAs in gene translation for quite some time now. More recently they have also demonstrated essential biological roles for numerous other ncRNAs, including microRNA , small interfering RNA and long ncRNA among others. All participate in gene expression regulation and thus serve as actors in organism development, environmental adaptation, etc. They are involved not only in biological processes but also in a number of diseases, including cancers and neurodegenerative pathologies. Notably thus, a better understanding of ncRNAs could contribute to a better understanding of these diseases and potentially to the development of novel therapeutics.

EvryRNA: close to twenty freely-available interactive tools for research

To energize this scientific field, the IBISC laboratory developed a wide range of bioinformatics tools designed to meet the needs of researchers and made them freely available to these latter via the EvryRNA platform. The more recent tools are built upon such technologies or algorithmic approaches as multi-objective optimization, self-organizing maps and deep neural networks among others.

Beyond ensuring performance, the IBISC researchers also sought to make their programs interactive and easy to use. Most thus offer numerous advantages:

  • an intuitive web interface;
  • an easily understood graphical representation of results;
  • the possibility for the user to modify parameters and re-run the model to improve its performance;
  • the possibility of including datasets or biologist knowledge in the model.

A platform accredited and supported by Genopole

Genopole supports the EvryRNA platform, which serves the biocluster’s innovative therapies strategic sector. Notably, Genopole financed a €60,000 informatics infrastructure update for a compute server and disk array.

The main programs available at EvryRNA

RANdvisor

RANdvisor enables integrated and holistic use of evaluation metrics for RNA 3D structure prediction. There are different metrics for evaluating the quality of RNA 3D structure predictions, but none perform sufficiently well alone. It is thus interesting to use several together.
👉 All metrics can be calculated via a simple line command.

DivideFold

DivideFold predicts the secondary structure (the first level of molecular folding due to base-pairing) of long ncRNA. Using a deep learning method, the program divides the initial sequence into smaller sub-sequences, which are then used for secondary structure prediction.
👉 DivideFold enables the study of long sequences, something that is difficult using literature tools.
👉 The sequence division can be re-trained on new datasets, which is a significant asset considering the low number of currently identified long ncRNAs.

RCPred

RCPred predicts secondary structures of multi-RNA complexes, i.e., conformations involving multiple, interacting RNA molecules. RCPred predicts the structure of each RNA molecule then searches for the minimum free energy structure for the complex.
👉 The tool is interactive: the user can act upon the different prediction steps by adding relevant information, choosing the most likely structures, etc.

C-RCPred

C-RCPred further optimizes RCPred predictions of multi-RNA complex secondary structures by integrating probing data (experimental data on certain base pairings) and user knowledge.
👉 Users can introduce knowledge and constraints into the system, thus better guiding its predictions. They can furthermore indicate corrections for the structure and re-run C-RCPred with those new data, and repeatedly so as many times as they wish.

IRSOM

IRSOM distinguishes coding and non-coding RNAs in a submitted sequence.
👉 Particularly, it is able to reject ambiguous sequences. The algorithm employs self-organizing maps. This type of neural network is advantageous in that it provides users with a map-like visualization of their results.

IRSOM2

IRSOM2 further exploits IRSOM’s ability to reject ambiguous cases to improve the identification of potential bifunctional RNA, i.e., those RNAs that comprise both a non-coding activity and a capacity for coding a functional protein.
👉 The tool gives biologists the option of autonomously retraining it using their own datasets and thus creating their own model. This aspect is particularly interesting considering the still incomplete knowledge on this class of RNA.

📍Accès à l’outil
🗞 Actualité sur IRSOM2

RNANet

RNANet is a pipeline for generating an integrated dataset for non-coding RNA, from the sequence to the 3D structure and including families, secondary structures and a large number of statistics.
👉 It seeks to be an ideal dataset for RNA machine learning.
👉 The dataset is updated monthly.

📍Accès à l’outil
🗞 Actualité sur RNANet

Biorseo

Biorseo provides secondary structure prediction using RNA modules (i.e., recurrent collections of ordered non-canonical interactions commonly found in RNA loops).
👉 Biorseo enables the efficacious prediction of RNA structures, including different motifs, e.g., pseudoknots, a characteristic RNA folding pattern.

BioKoP

BioKoP predicts RNA secondary structures with pseudoknots.
👉 Its particularity is to combine two prediction models, which increases the likelihood of generating a structure as close as possible to the real one.

miRNAFold

miRNAFold seeks to identify microRNA precursors (pre-miRNAs) at a large scale in genomes.
👉 A clear and easy-to-use web server.
👉 Can treat entire chromosomes.

miRBoost

miRBoost enables the classification of pre-miRNA candidates as true or false pre-miRNAs using a machine-learning method (support vector machine).
👉 The model can be retrained on new data.

IpiRId

IpiRId predicts piwi-interacting RNA (piRNA) via a holistic approach combining numerous functionalities in a machine learning method, specifically multiple kernel learning. This latter involves defining a kernel per feature, which makes for a method that is both adaptive and extendable according to the user’s dataset.
👉 A web server that allows the user to choose the data to be considered.
👉 Possibility for the user to retrain the model.

MORE ON THE PLATFORM OFFER: A GENOPOLE SPECIFICITY


Genopole ensures access to 24 shared-use technological and ancillary services platforms for its businesses and laboratories.
Its objective is to provide these entities with concrete solutions across all biotech-associated fields of research: cellular biology and imaging, molecular biology, structural biology, bioproduction, biological resources, functional testing, bioinformatics and robotics & automation.

Counting collectively more than 650 shared-use devices, some of which represent the cutting-edge of technology, Genopole’s platforms are key enablers of success, empowering novel discoveries for the biocluster’s academic labs and confirmation of the potential of innovations forwarded by its companies.

@: Julien.Picot@genopole.fr

Share
With the support from
Région île de France