KNOWLEDGE FROM DATA AND ACTION FROM INSIGHTS
The PercayAI Software Platform was designed to augment hypothesis generation, target identification, and dry lab analysis to go from insight to action faster and with more confidence.
The PercayAI Software platform was developed to rapidly accelerate the often difficult and time consuming interpretative and hypothesis generation efforts that accompany large biological data sets.
Created by a small group of computational biologists with broad disease domain expertise and over 50 years combined experience interpreting omics data to drive decision processes in industry and academic environments.
PercayAI tools were designed to overcome many of the shortcomings common to tools in this space and to algorithmically mimic the interpretative thought processes of domain experts, thereby reducing the time to hypothesis and/or conclusion by an order of magnitude or more.
Limitations of exsisting tools
EXISTING TOOLS PROVIDE SOME INSIGHT - WITH SIGNIFICANT LIMITATIONS
Most current tools designed to aid in analysis and interpretation of omics data fall into two major classes, overlap/enrichment and natural language processing. While quite different in mechanism, both can provide some degree of insight but with substantial limitations.
Some of these limitations are common, such as the need for existing definitions such as ontologies or pathway/gene associations, while others are unique to the class but both limit what can be identified. Most importantly, all lack the ability to provide a “big picture” overview of the data.
Most limited to single omics type
All lack "big picture" view
PATHWAY & PROCESS OVERLAP/ENRICHMENT
• Many output lists required to cover all facets
• Lists tend to be highly redundant
• No interconnectivity within or across lists
• Creates structured data to return assertions
• Samples the “conscious” (bias) of writers
• Context is lost; Localized associations
What makes PercayAI software different?
SO IF YOU COULD...
Assemble Information from Numerous Knowledge Bases for Multiple Omics Types
Remove Highly Redundant Information
Interconnect All Related Information
Display as a Holistic Interactive Map
YOU WOULD HAVE SIMILAR TOOLS TO THE PERCAYAI PLAFORM
As highly-experienced translators of large-scale omics data, the developers of PercayAI sought a platform that could that could accelerate the discovery and hypothesis generation process. Through its unique design, the platform addresses many of the knowledge assembly and assimilation limitations of existing methods, thus enabling rapid interpretation.
HOW DOES THE PERCAYAI PLATFORM WORK?
Identifies the entire literature corpus associated with the input entities
Contextual Language Processing (CLP) identifies concepts enriched in this corpus
Inter-related concepts form biological themes
Themes are displayed in 3D with spatial relationships and inter-connectivity
Samples the “subconscious” of literature specific to entity list
The PercayAI suite enhances the knowledge assembly process through the use of Contextual Language Processing (CLP) in and ontology-free manor. CLP recognizes the fact that the meaning of a concept as well as the relevance of gene, protein, metabolite, or other biological entity may differ with context. Thus, CLP engine breaks language down to the “atomic” level of words and then reassembles those words as concepts and then themes in a contextually relevant framework.
Assembly of information represents only half of the challenge in the interpretation of complex biology. If that information cannot be absorbed by a researcher in a meaningful and helpful way, it never becomes knowledge. With this in mind, the platform was developed with an advanced, intuitive visualizations to enable rapid assimilation of complex information.
The PercayAI plaform vastly reduces data complexity and identifies explicit and implicit associations in an ontology-free and context-dependent manner. The visualization framework allows researchers to see the big picture view of their data but allows them to “drill up” or “drill down” at numerous points of the analysis. The unique search engine provides a means to search and see relevant images related to concepts within a theme that can provide conceptual insights for biology that may be unfamiliar to the researcher.
ASSEMBLES INTERPRETABLE KNOWLEDGE, NOT LISTS
In current tools, the output of a single analysis may be as overwhelming as the input lists.
Displays a unified and interconnected view of contextually-relevant themes
Identifies important unique findings