Developed by Alberto Paccanaro and his team, LanDis is freely available online, providing support for differential diagnoses

Alberto Paccanaro conducts research leveraging AI and Machine Learning techniques to address challenges in Systems Biology, Medicine, and Pharmacology | Photo: FGV EMAp

Medicine and biology are undergoing a revolution driven by data integration and technology. At the forefront of this transformation is Professor Alberto Paccanaro from the School of Applied Mathematics at Fundação Getulio Vargas (FGV EMAp), with the launch of LanDis, an interactive tool that significantly contributes to the analysis of hereditary diseases.

LanDis leverages the concept of the interactome—a network representing molecular interactions within our bodies. Diseases that share genes tend to cluster in specific regions of this network, known as disease modules. Disruptions in these modules can lead to illnesses, and diseases with similar clinical features often localize near each other in the interactome.

Globally accessible and open access, LanDis enables doctors and researchers to explore over 44 million pairs of hereditary diseases, facilitating their comparison and supporting differential diagnoses. “We can think of LanDis as a tool that helps identify which genes might be responsible for a hereditary disease, much like a machine diagnosing issues in its components. By predicting which genes may cause a hereditary disease, it becomes possible to consider treatments that address the root cause of the problem”, explains Paccanaro, who holds a PhD in Computer Science from the University of Toronto.

Paccanaro and his team’s work on LanDis was recently published in the article titled "LanDis: the disease landscape explorer" in the European Journal of Human Genetics, part of the Nature group, highlighting the tool's significance in advancing research on hereditary diseases.

Alberto Paccanaro (wearing a striped shirt) with his team of researchers from FGV EMAp, responsible for developing LanDis, an innovative tool for hereditary diseases | Photo: FGV EMAp

How does LanDis work?

The platform employs a technique called Caniza similarity, also developed by Paccanaro and his team a few years ago. This method gathers information from scientific publications using medical annotations known as MeSH (Medical Subject Headings). These data are used to calculate the similarity between diseases, enabling scientists to identify patterns that might not be immediately evident. In this way, LanDis can highlight important connections between diseases that share causative genes.

A practical example is Tetralogy of Fallot (TOF), a congenital heart disease. In LanDis, TOF appears at the center of a cluster of related diseases, such as Alagille Syndrome (ALGS1) and Right Atrial Isomerism (RAI), which share genes and exhibit similar symptoms. This visualization aids in discovering new genes that may be associated with hereditary diseases that are still underexplored.

LanDis chart showing Tetralogy of Fallot (TOF) connected to other diseases. The lines indicate genetic similarity, aiding in the identification of connections between hereditary diseases | Image: Reproduction.

These predictions could open doors to new discoveries about the origin of hereditary diseases and assist in the development of personalized treatments. "We are using this information to collaborate with teams from around the world, such as researchers from the United Kingdom investigating the link between heart diseases and rheumatoid arthritis," adds Paccanaro.

Integration with large databases

In addition to its visualization tools, LanDis is integrated with renowned databases such as OMIM (Online Mendelian Inheritance in Man), which catalogs genetic diseases, and UniProt, which provides detailed information on proteins. These integrations allow researchers to access essential molecular data directly through the platform.

"The system was developed to be accessible and easy to use, without the need for specialized equipment," emphasizes Paccanaro.

The source code of LanDis is available for free on GitHub, allowing other researchers around the world to use and enhance the tool. The platform has been tested on major browsers and operating systems, and it works optimally on Google Chrome.

A researcher from the Paccanaro Lab analyzes molecular models in computational simulation | Photo: FGV EMAp

A tool for the future of Medicine

LanDis offers a new perspective for exploring the relationships between diseases, enabling more accurate differential diagnoses and helping scientists focus on the clinical features most critical to each condition.

The ability to analyze large volumes of data is what makes LanDis a powerful tool. "The medicine of the future will increasingly be connected to the analysis of large data sets, and LanDis is an example of how we can use this information to improve the treatment and prevention of hereditary diseases," concludes the researcher.

"Today's biologist needs to interpret terabytes of data to answer their questions, and this is where data science and machine learning come in," emphasizes Paccanaro.

The projects at PaccanaroLab are funded by various institutions, including Fundação Getulio Vargas, the National Council for Scientific and Technological Development (CNPq), the Carlos Chagas Filho Foundation for Research Support of the State of Rio de Janeiro (Faperj), the Medical Research Council (UK), and the National Science Foundation (USA).

A researcher from the Paccanaro Lab reviews mathematical equations for modeling complex networks | Photo: FGV EMAp

For more information about LanDis and other projects from PaccanaroLab, visit paccanarolab.org/landis.