Behind a user-friendly interface, a complex and powerful artificial neural network

DeepProphet: The state-of-the-art machine learning techniques serving Life Science Research

GeneRecommender utilizes a proprietary neural network known as DeepProphet, which analyzes an extensive collection of scientific articles and processes approximately 445 million connections. By inputting a selection of genes and/or diseases of interest, this engine generates recommendations for additional genes that are closely associated and should be taken into consideration by users. To leverage the latest information, we have integrated data from multiple public databases.


GeneRecommender's Artificial Intelligence engine has been trained using an extensive dataset comprising millions of scientific papers. Throughout the learning process, the Artificial Neural Network analyzed the collective contributions of life science researchers worldwide over the past three decades. By extracting patterns and correlations among genes and diseases, GeneRecommender's AI engine acquired valuable insights from this vast body of knowledge.

Scientific Papers Scanned
GeneSymbols Considered
Diseases Considered
Genes Citations
Diseases Citations



Artificial intelligence (AI) is a wide-ranging branch of computer science concerned with building smart machines capable of performing tasks that typically require human intelligence. Within the concept of AI, different methods and techniques can be found. Nowadays, one of the most useful and promising of these techniques and methods is Machine Learning.


Machine Learning is a multifaceted notion that draws on methodologies from statistics, computer science, and data processing. Machine Learning products are computer programs that use data to formulate predictions, get accurate answers, and make decisions without being explicitly programmed, but rather learning from data as humans do daily.


Classical software engineering involves an expert setting precise rules when developing software. In many cases formulating the rules can be prohibitively complex. Machine Learning algorithms work in a different way. All patterns discovered in the past would draw new conclusions to form new data sets. This makes it possible to process a really huge amount of information.

The Algorithm - DeepProphet2

GeneRecommender uses DeepProphet2, an advanced deep learning algorithm trained on all the scientific papers available online. Indeed, using cutting-edge Natural Language Processing (NLP) techniques, the system extracts key information from manuscripts and uses it to train a deep learning neural network to predict which genes might be associated with a given study. In order to accomplish this task, TheProphetAI developed a customer transformer-based model based on a design similar to that of the most famous and successful language models, such as LaMDA and GPT-3. We have tested the System using gene sets representing pathways and diseases. Recursively, a randomly chosen gene was left out and the remaining input was provided, we tested whether the algorithm could predict which gene was missing based on its first recommendation. Since the algorithm does not know the gene sets during training, if it can complete them, it has learned an accurate representation of genes. Using this kind of analysis, we reached an astonishing result: an AUC of 0.9.

If you would like to go into technical details about the algorithm and its validation, please consider downloading our scientific paper: []

Human and AI cooperation

The integration of Artificial Intelligence systems into your workflow can bring about significant advantages. To fully unlock the potential of these technologies, it is important to adopt AI in a safe and appropriate manner. While AI cannot replace humans in their roles, it can provide valuable assistance and support in daily tasks, enhancing efficiency and effectiveness. Although AI predictions are not always infallible, they offer a valuable starting point for addressing specific issues.

We firmly believe that the true potential lies in the collaboration between humans and machines. In-depth analysis by experts is necessary to comprehend the actual value of the recommendations derived from the AI engine. While our platform offers tools to comprehend the outcomes of each recommendation, it should be noted that we rely on AI precisely because working with such vast amounts of data is not feasible for humans alone.

Are the results always accurate? While perfection may not be attainable every time, a sufficient number of targets can yield statistically valuable results. Most of the recommended genes will provide useful insights for your research.

The process followed by Artificial Intelligence mirrors that of humans, but it operates at a fraction of the time. Researchers, like yourself, often refer to scientific literature to understand previous work on similar topics and gain deeper insights into their own research. In fact, we employ the same data, processed by a powerful neural network that possesses knowledge of the scientific community’s prior studies and experiments, enabling it to generate recommendations based on this wealth of information.

AI will not replace your work, but it will expedite your studies by providing innovative and indispensable insights.

Data sources included in the Platform

Griss J, Viteri G, Sidiropoulos K, Nguyen V, Fabregat A, Hermjakob H. ReactomeGSA - Efficient Multi-Omics Comparative Pathway Analysis. Mol Cell Proteomics. 2020 Sep 9. doi: 10.1074/mcp. PubMed PMID: 32907876.

Our Development Values

Two-cell embryo, Mitosis under microscope (3D illustration)



Security and safety

User Oriented

We are committed to real innovation

TheProphetAI develops its algorithms and models to bring innovation to the life science sector. Machine learning, artificial intelligence, and deep learning are not just buzzwords for us. They are the foundation of our platform.

TheProphetAI performs rigorous tests on every algorithm

Platform algorithms have all been tested against public benchmarks. Indeed, we perform rigorous tests on every model you may find on GeneRecommender. If you would like to know more about the performances of our AI model please refer to our paper.

The platform has been built following the highest security and safety standards

We believe in the importance of privacy. Data are “the new gold” and for this reason, should be managed in the most careful way. The platform itself has been built in order to protect your data.

The complexity of a neural network with an easy to use interface

Artificial Intelligence systems, Machine Learning and Artificial neural networks are extremely complex mathematical and statistical models. Unfortunately, these techniques are as complex as hard to implement. GeneRecommender will allow you to get the benefits from the AI without having to deal with its complexity, thanks to an easy-to-use interface, with no need for coding skills.

  • Innovation
  • Robustness
  • Security and safety
  • User Oriented