Artificial intelligence is a crucial tool for discovery in the Health and Life Sciences. Its application allows biotechnologists to collect, classify, and analyze an unprecedented amount of biochemical and biological information. AI is also a powerful generative tool, that can be used by biotechnologists to create high-fidelity, real-world data when suitable data is too sensitive, biased, or unavailable. At Lynntech, we harness both aspects of AI to successfully transform ‘omic’ data into actionable knowledge.
Capabilities
Forensic genomics utilizes modern genetic testing and genomic analysis techniques to extend the type of information provided by a forensic DNA sample. In addition to confirming identity, a forensic genomics approach can identify biogeographic ancestry and kinship of a DNA sample, and with advancing technology, even build a phenotypic profile to predict appearance. Lynntech is harnessing the power of AI to develop a forensic genomics stack that meets the needs of the DOD and law enforcement agencies, establishing the foundational validity of the approach, and demonstrating its utility in real-world applications.
Recent advances in AI have significantly decreased the time and increased the fidelity of model-driven bioinformatic studies. AI can be used to predict protein structure and function, and these advances will, in the future, be expanded to the design of novel synthetic proteins and enzymes. AI, in combination with personal genomics, is powering the ability to offer precision medication treatment plans; thus, improving patient outcomes and reducing risks. Finally, AI-driven metabolomics is being used to rapidly predict the biotransformation of original chemicals into novel metabolites, reducing biopharmaceutical development time.
AI is a potent tool that is increasingly finding applications in the administration of healthcare. These computationally powerful algorithms are especially suited for complex pattern recognition in large datasets of images. Lynntech aims to develop robust AI algorithms for creating diagnostic modalities to detect disease/injuries using fundus photographs, optical coherence tomography and visual fields. The long-term goal is to develop AI-driven diagnostic platforms that can be used for screening purposes at point-of-care facilities for early detection of disease and referral to specialists. Such a healthcare delivery model can: (1) reduce the incidence of adverse outcome by ensuring earlier onset of care by a specialist, and (2) lower healthcare costs by avoiding unnecessary medical intervention in patients who are unlikely to benefit from such measures.
Lynntech aims to develop AI algorithms for rapid identification of novel candidate molecules that are clinically safe and therapeutically active. Our approach is based on AI and machine learning (ML) algorithms that are coupled with structural and mutational analysis data to create virtual libraries of infinite space. Such a strategy allows for an iterative implementation to rapidly identify and evaluate new drug candidates for a variety of clinical indications e.g., mounting a rapid response to emergence of novel multidrug resistant bacterial strains.
Selected Project: DNA Forensics
Many real world problems cannot be easily solved using direct modeling and conventional imperative programming. Lynntech uses AI methods in combination with state-of-the-art electro-optical/infrared devices, embedded navigation systems, human-machine interfaces, and other resources to solve real-world problems such as maritime search and rescue, analysis of geospatial data to provide actionable intelligence, terrain monitoring, disaster recovery, computer vision for autonomous vehicles, assistive technologies for disabled users, data analytics, natural language processing, and mobile computing for Internet of Things (IoT) applications.
Operational Need – Forensic genomics is an under-utilized tool for military applications that combines DNA sequence data with computational approaches to provide actionable intelligence. Pedigree reconstruction dramatically expands the capability of forensic DNA identification by inferring the relationships of individuals far beyond those that are currently present in a DNA database. Reconstruction of family lineages has multiple military applications, including establishing the identity of war casualties from fifty-plus years ago and the identification of familial lineages within terrorist networks. In both cases, stringent statistical methods are required to provide actionable intelligence. A current limitation of the use of forensic genetics, in particular kinship and phenotypic prediction, is the black box nature of many of the proprietary approaches currently in the market.
Lynntech Solution – Lynntech proposes Quick Single Nucleotide Polymorphism (Quick-SNP) as a DNA forensic genomics tool. Lynntech’s Quick-SNP will analyze naïve DNA profiles, and in the process, establishes sample identities, predicts familial relationships (kinship and biogeographic ancestral), and ultimately provide phenotypic profiles. To establish the foundational validity that the techniques are scientifically sound, replicable, and accurate in a lab environment, and to ultimately demonstrate applied validity in real world applications, Quick-SNP will follow the guidelines established by the President’s Council of Advisors on Science and Technology for the Forensic Sciences (Lander, 2016).
Revolutionary Performance – Lynntech has selected a core group of SNPs across the genome to fingerprint individuals, reveal their biogeographic ancestry, family kinships, and with advancing technology, even build a phenotypic profile to predict appearance. Lynntech has optimized Quick-SNPs algorithmic and computational architectures to decrease the processing time by >100X compared to classical, non-parallelized methods. Quick-SNP utilizes a statistical approach that increases the stringency of the results, reduces the search space, and increases the depth of pedigree prediction.