Your smart software solutions for cheminformatics and QSAR research.

About Alvascience

Founded in 2018, Alvascience is striving to boost value for cheminformatics research through high-tech software solutions using machine learning and other state of the art science tools.

At Alvascience, we are constantly exploring and implementing the most promising and innovative technologies in our software tools, which makes them a leading choice for QSAR and other cheminformatics research. By providing time-, cost-, and labor-efficient solutions, we are benefiting scientific and industrial researchers across pharmacological, cosmetic, agri-food, materials science sectors, and more.

About Our Products

Our software tools use in silico techniques to analyze chemical datasets and to evaluate the physico-chemical and ecotoxicological properties of chemicals, which helps reduce the need for in vivo and in vitro experimentation.

Our solutions support researchers throughout their work, to generate, handle, and analyze molecular datasets in silico. Starting from data curation (alvaMolecule), the calculation of molecular descriptors and fingerprints (alvaDesc), regression and classification modeling (alvaModel), the deployment of the models (alvaRunner), to de novo molecular design (alvaBuilder), our products can be used independently or as a suite of tools to tackle QSAR workflow.

alvaMolecule

The tool to view and prepare your chemical dataset.

alvaMolecule is a molecular worksheet made for visualizing, analyzing, curating, and standardizing your molecular dataset.

With the help of checkers and predefined standardizers, alvaMolecule covers the whole data curation process.

Learn more

alvaDesc

The tool to calculate molecular descriptors and fingerprints.

As the next-gen tool, alvaDesc calculates thousands of molecular descriptors including molecular properties, drug-like, and lead-like indices.

With innovative features and various auxiliary tools, alvaDesc simplifies and innovates your calculation and analysis process.

Learn more

alvaModel

The tool to create QSAR/QSPR regression and classification models.

Based on Genetic Algorithms, alvaModel performs a feature selection to find the best models according to a defined score.

These models can be used to predict the biological, physicochemical, and environmental properties of chemicals.

Learn more

alvaRunner

The tool to apply QSAR/QSPR regression and classification models.

alvaRunner calculates the descriptors and fingerprints to apply the given QSAR/QSPR regression and classification models without the need for any other software tool.

With a user-friendly graphical and command-line interface, alvaRunner makes efficient model application a reality.

Learn more

alvaBuilder

The tool for de novo molecular design.

alvaBuilder is specifically designed for generating novel molecules with a desirable set of properties – starting from a training set of your choice.

With its simple user interface and step-by-step procedure, your de novo molecular design is easy to manage at your fingertips.

Learn more

Alvascience at a glance

Companies

Alvascience tools have been used by more than 70 companies across industries around the world, including pharmaceuticals, electronics, agrochemical and materials science companies.

Academics

Alvascience solutions have been used in more than 150 universities and research centres in more than 30 countries assisting researchers in their cheminformatics projects.

Citations

Alvascience tools have been cited in more than 300 papers published in peer-reviewed scientific journals exploring solutions from drug discovery to materials science.

Our satisfied clients

alvaDesc is a neat tool to generate molecular descriptors and fingerprints. We have been using it in many of our preclinical QSAR modeling work for generating in silico models of ADME parameters. So far, we have noticed that the resulting QSAR models show really good predictivity and in some cases, the predictivity is very similar to predictions from actual in vivo data when testing new compounds that are not in the training set. These models are now our internal recommendation.

Shibin Mathew

Principal Scientist, Pfizer

alvaDesc is an extremely useful piece of software for conducting material exploration using machine learning. It has more than 5,000 features that are necessary to build a prediction model from data, so it is possible to build a prediction model with extremely high accuracy. The software can be used on Windows and Linux and is extremely easy to operate. Based on the above, we recommend this software as a useful software.

Michihiro Okuyama

Principal Data Scientist, Konica Minolta

alvaBuilder is a great addition to drug design pipelines. The friendly user interface lets you start your de novo design immediately. I recommend it because of its fast algorithm while managing large datasets. Also, the possibility of easily changing training sets and the scoring function is useful to work on different projects without fragmenting databases each time. The availability of different descriptors to set the rules of the scoring function also comes in very handy.

Diana Lorena Prado Romero

PhD Student at National Autonomous University of Mexico (UNAM)

alvaDesc is one of the most powerful tools for calculating a wide range of molecular descriptors and fingerprints. The quality of the alvaDesc descriptors is exceptional while being easy to use. It also has the advantage of being well integrated with tools such as KNIME. We have been using alvaDesc for over 4 years with outstanding results in four of our QSAR articles. We definitely recommend alvaDesc as a tool for generating descriptors and fingerprints in the field of computational chemistry.

Christophe Molina

CEO, PIKAÏROS

In our daily work as computational materials scientists, alvaDesc stands as a valuable tool to quantify molecular structures. Its diverse range of molecular descriptors, easy usage, compatibility with other software and its Python interface make it a versatile choice. Due to alvaDesc, we were able to significantly boost the predictive performance of our machine learning models in the design of corrosion inhibitors. Strongly recommend!

Tim Würger

Postdoctoral Researcher at Helmholtz-Zentrum Hereon

We are very happy to use alvaDesc in our research projects. Has a lot of nice and useful features! Highly recommended!

Jose L. Medina-Franco

Full Professor, National Autonomous University of Mexico (UNAM)

In the last few years, we have been users of alvaDesc software and the features it provides, for its application in the field of chemoinformatics are incredible. Its implementation in KNIME has also been very relevant and increases its potential use. Thanks to the developers for this software that they have put in our hands!

Miguel Ángel Cabrera-Pérez

Scientific Consultant, PKIT Consulting

We find alvaDesc a nice tool for our polymeric materials analysis. It supports the calculation of more than 5,000 descriptors, which are quite helpful in cheminformatics studies to understand the structure-property relationship of molecules. A nice tool to explore!

Lei Tao

Postdoctoral Research Associate, University of Connecticut

AlvaRunner is to our knowledge the best software available for predicting biomagnification factors (BMF) for hydrocarbons. The predictions are in good accordance with experimental data. The alvaRunner KNIME node is a great tool for dealing with big datasets!

Yves Verhaegen

Ecotoxicology Research Associate at Concawe

I have been using alvaDesc at Pikairos for around 6 months now to generate Physchem descriptors and fingerprints for use in ADMET predictive modelling. I use the desktop software and also the KNIME nodes. I have found that alvaDesc quickly generates a large number of good quality descriptors that have proven to be useful with multiple machine learning methods. I can highly recommend using this software! Thank you to the Alvascience team for such a great tool!

Heather Lambert

Cheminformatician at Pikairos

I do recommend alvaMolecule software due to its ability to handle large datasets (curation and standardization) efficiently and because it provides various tools, including several useful molecular descriptors. It is also very user friendly.

Lorella Spiteri

PhD in Computational Crystallography

[FREE] New alvaRunner model: Prediction of Bioconcentration Factor (BCF) of chemicals

Bioconcentration Factor (BCF) model

10.12.2024

[FREE] New alvaRunner model: Prediction of Bioconcentration Factor (BCF)

Prediction of BCF of chemicals

Contact us

If you are interested in finding out more about Alvascience or any of our products and services, please do not hesitate to contact us.

Together, let’s create more value in cheminformatics!