Biovista

Infobox Company
company_name = Biovista, Inc.
company_type = Private
company_slogan = To Seek, To know, To Act
foundation = Charlottesville, VA, USA (2005)
location = flagicon|USA Charlottesville, VA, USA
industry = Biotechnology
services = Drug repositioning
homepage = [http://www.biovista.com/ www.biovista.com]

Biovista Inc. is a private drug development services company based in Charlottesville, Virginia, USA. Biovista's core business activities include Drug repositioning and drug de-risking as well as disease cohort analysis, adverse event prediction and clinical hold analysis services. Biovista is also applying its technology platform to develop its own drug repositioning programs in the areas of CNS, diabetes/obesity, eye disorders, and oncology.

The company derisks and repositions drugs using multidimensional profiles of pharmacologically relevant entities such as genes, diseases, drugs, pathways and cell types, to identify and rank potential adverse events and new indications for drugs in development, on the market, or generics.

Biovista is also creating software-based tools and services for , researchers in the Life sciences and the consumer & patient health areas.

__TOC__

Technology Platform

Biovista's technology platform is based on the analysis and integration of Biomedical information available in the scientific literature using Biomedical text mining techniques. Pharmacologically-relevant areas include drug toxicity, drug mode of action, disease mechanisms and biological system interactions.Biovista Inc.’s technology platform integrates literature-based discovery algorithms with Semantic search technologies to identify and rank potential solutions to a variety of drug development related problems such as predicting the adverse events of compounds, identifying suitable biomarkers for diseases and discovering new indications for existing drugs or drug combinations.

Biovista’s correlation engine scans potential interactions between pharmacologically relevant entities resulting in a correlation database. The database itself is based on a proprietary design that combines the RDBMS model with the Object-Oriented model allowing researchers to obtain preliminary answers in weeks rather than years.

Because of its predictive nature, Biovista’s technology platform has been evaluated in order to assess its predictive accuracy, predictive lead times and robustness. Results from the evaluation of American Society of Clinical Oncology (ASCO) announcements show accuracy in the region of 65% with predictive lead times of 5-8 years. While the output from the discovery engine requires additional assessment by subject matter experts, these kinds of results suggest that a literature based approach that can indicate adverse events 5 years before they are officially reported could offer a significant aid to pharmaceutical companies and regulatory bodies (such as the FDA and EMEA) alike.

Research

Biovista is an active participant of European Union co-funded R&D projects spanning areas such as post-genomic clinical trials research (ACGT project), mutant mouse models for the investigation of Human Immunological Disease (MUGEN project), semantic annotation and ontology driven text mining (PARMENIDES project) and systematic knowledge discovery (ESPERONTO Project).

External links

*Biovista Inc. – Home http://www.biovista.com
*ASCO [http://www.asco.org/]
*US Food and Drug Administration (FDA) [http://www.fda.gov/]
*European Medicines Agency (EMEA) [http://www.emea.europa.eu/]
*MUGEN European Project [http://www.mugen-noe.org/]
*ACGT European Project [http://www.eu-acgt.org/]
*PARMENIDES European Project [http://www.ifi.uzh.ch/cl/Parmenides/]
*ONTOWEB European Project [http://www.ontoweb.org/]
*ESPERONTO European Project [http://webode.dia.fi.upm.es/sew/esperonto.html]

References

# [http://maillists.uci.edu/mailman/public/mgsa-l/2006-March/006781.html Creating order out of information chaos (Parmenides project)]
# W J Black, L Gilardoni, F Rinaldi, and R Dressel. Integrated text categorisation and information extraction using pattern matching and linguistic processing. In Proceedings of RIAO97, pages 321–335, Montreal, 1997.
# [http://www2.informatik.hu-erlin.de/Forschung_Lehre/wm/ws04/10.pdf Fabio Rinaldi et al Mining relations in the GENIA corpus] , In Proceedings of the Second European Workshop on Data Mining and Text Mining for Bioinformatics, Pisa, Italy. 24 September 2004.
# F . Rinaldi , G . Schneider , K . Kaljurand , M . Hess , C . Andronis , O . Konstandi , A . Persidis Mining of relations between proteins over biomedical scientific literature using a deep-linguistic approach☆☆☆ . Artificial Intelligence in Medicine , Volume 39 , Issue 2, Pages 127 - 136
# [http://csdl2.computer.org/persagen/DLAbsToc.jsp?resourcePath=/dl/proceedings/wi/&toc=comp/proceedings/wi/2006/2747/00/2747toc.xml&DOI=10.1109/WI.2006.128] 2006 IEEE/WIC/ACM International Conference on Web Intelligence (WI'06) pp. 1057-1060 PARMENIDES: Towards Business Intelligence Discovery from Web Data. Alexander Mikroyannidis, Babis Theodoulidis, University of Manchester, United Kingdom, Andreas Persidis, Biovista, Greece
# [http://www.hum-molgen.org/companies/profile.php3/204 HUM-MOLGEN Registry of biomedical companies]