Our Technology - Iktos
17254
page-template,page-template-full_width,page-template-full_width-php,page,page-id-17254,ajax_updown_fade,page_not_loaded,,qode-title-hidden,qode_grid_1300,qode-content-sidebar-responsive,qode-theme-ver-10.1.1,wpb-js-composer js-comp-ver-5.0.1,vc_responsive

A proprietary module generator

Deep Learning Generative Models

Generative models are new algorithms based on deep learning technology which enable to generate new data points…

Learn more

Combining state of the art computational chemistry and machine learning technologies to solve the issues of new compound design

 

Deep learning Generative models

Generative models are new algorithms based on deep learning technology which enable to generate new data points. After having learned a task (create a picture, write music, generate new content…) the algorithm becomes capable of imagining new images, pieces of music or texts.
Based on this approach and using publicly available databases as an input, Iktos algorithm has learned to design new molecules.

 

Objective functions

To guide the molecular generator in the infinity of the chemical space, we have introduced fitness functions that the algorithm must optimize. Those functions are predictors, specifically built from your project’s data set: your molecules, measured on your assays, activity, selectivity, ADMET… On top of that we also have implemented fitness functions to maximize druggability, synthetic access, and similarity to your initial data set. The molecules we generate are optimized simultaneously on all your success criteria.

 

Top notch predictive models

Iktos molecule generation technology is only as good as the predictors it uses to travel in the chemical space. The quality of the predictors is key.
We therefore have developed ways to construct high-performance predictors and to select the optimal ones, automatically and in a few hours, by benchmarking and optimizing both machine learning algorithms and molecules descriptors (1D, 2D and 3D representations of the molecules, and even data driven representations like latent space for instance).

A tailor made IT infrastructure

A highly powerful infrastructure

Iktos has developed its own big data plateforme to be able to parallelize GPU and CPU on demand…

Learn more

A Tailor-made IT infrastructure

 

A state-of-the-art big data infrastructure

Iktos has developed its own big data platform that enables to parallelize computation on GPU and CPU on demand, and can easily be scaled up to meet any computation needs.

 

Molecular structures generation metrics

Our monitoring tools enable real-time supervision of the process and visualization of the results as they are generated:
– Speed of convergence to the desired objective
– Molecular diversity of the generated structures
– Similarity to the initial data set
– Etc…

We are different

Virtual screening

Virtual screening has many well-known limitations that limit its usefulness in practice. In contrast, Iktos technology is a “virtual designing” algorithm that designs molecules …

Learn more

We are different

 

Virtual screening has many well-known limitations that limit its usefulness in practice: time consuming, strong expertise required, dependence upon fragments libraries and parameter setting, limitation to a small number of targets…

In contrast, Iktos technology is a “virtual designing” algorithm which is guided to create novel molecular structures that are optimized to meet any given multi-objective blueprint.

 

Traditional approach: virtual screening

• Need to define a fixed scaffold and allowed structural modifications at selected places
• Combinatorial approach, dependent upon fragment databases availability
• The size of the database must increase exponentially with the number of objectives
• Need a high amount of data preparation, computation time and expertise

 

IKTOS: A virtual designing technology

• No prerequisites on the scaffold, the algorithm identifies what drives biological activity
• No dependence on virtual molecular fragments databases
• Thanks to our proprietary algorithm molecular diversity is limitless
• The number of parameters to optimize does not impact performance
• A few hours only are needed to identify promising molecules