Transcell Biolife - Your next generation Biobank

Hyderabad based Transcell Biolife (www.transcell.in) partners with Bangalore’s Institute for Applied Research and Innovation https://www.instari.org/ to deliver Genomic intelligence based deep-learning for better healthcare through integrated biobanking solutions – An academic partnership extrapolated to redefine personalized medicine

Artificial intelligence (AI) or cognitive intelligence (CI) is the talk of town now days but it is not just one final destination, it needs to be utilized in a way for better life and that’s where application of intelligence is important. Genomics is a field where large amounts of data are generated and evaluating, identifying biomarkers (predictive and prognostic) and their application for improved healthcare are important. In specific, when we talk about incorporating genomics into healthcare, establishing credible intelligent platforms using data generated in laboratories plays a major role. These intelligent platforms derived from large scale genome studies with complicated data from DNA sequence to patient/donor information are needed to develop computational models. Dr. Rajani Kanth Vangala, Chairman & Founder of Institute for Applied Research and Innovation (InARI) has more than 10 years of experience in running one of the largest clinical studies called Indian Atherosclerosis Research Study (IARS as a director of Thrombosis Research Institute) where they collected 13000 patients data and clinical samples (whole blood, serum, plasma, saliva, urine and blood vessel tissues). He led the team in creating genomic and proteomic biomarkers based risk prediction algorithm for coronary artery disease (CAD) using CI.

The CI based tools can augment and extend the effectiveness of a physician’s practise. With more data and information out about whole-genome sequencing and biometrics, there are more expectations from patients/donors to get holistic insightful and personalized care. Physicians have always been in need of identifying and interpreting relationships between different variables for improving patient care. CI and machine learning use several methodologies which allow algorithmically learning and efficiently representing the data output. The difference between usual statistical methods used by physician and machine learning is that with former one can conduct inferences about given set of samples or population but with later, we can predict or develop classification approaches. However, these two approaches are intertwined and are used together to develop applied platforms.

Deep-learning is a subfield of machine learning where large data driven rules are automatically used to improve diagnosis or therapy. This process of machine learning happens by feature extraction from raw data and using these features as learning algorithm to detect patterns in new inputs. In contrast, the deep learning develops its own pattern recognition approaches using multiple hidden layers of sequentially arranged primitive, non-linear representations or features. As data flows through these layers, the system iteratively identifies the patterns and associations. Deep learning models are usually large scale datasets with ability to run on specialized computing hardware and continuous improvements. Convolution neural networks (CNNs) are a type of deep-learning methods which can process data with spatial invariance like images. Image-based diagnostics has successfully implemented CNN-based methods. Genomics is the field where deep-learning approaches are adapted beyond conventional approaches like NLP (Natural Language Processing), CNN and RL (Reinforcement Learning). The genomic technologies can output wide varieties of data from an individual’s DNA sequence to quantity of functional proteins in different tissues in different medical conditions. These kinds of results help clinicians to provide more accurate diagnostics and treatments if embraced.

A usual deep-learning genomics pipeline starts with raw data (example: DNA sequence, gene expression etc..), converts this raw data into input data tensors which in turn are fed through neural networks to power the biomedical needs. One of the major opportunities is large-scale genome-wide association (GWA) studies which aim at discovering causal or associated genetic markers. Applying such algorithms for GWA studies in large patient/donor cohorts can help in dealing with latent confounders. Understanding the genetic modifications can allow the clinicians to recommend treatments and based on associated factors. One major step ahead for deep-learning is integrating different external data sources into GWA studies like medical imaging or other phenotypes result in improved identification of disease associated causal mutations. However, for better understanding and easier utilization of the genomic intelligence, data visualization tools are important.

Intuitive visualization tools are needed in order to effectively interpret and extract new biological knowledge and insights, more so integrating the previous biological and clinical knowledge.

Therefore, it is important to access large repositories of often structured biological knowledge and facilitate interaction among them. These kind of tools and deep-learning are revolutionary in healthcare which could translate into great benefit to practicing clinicians.

This collaboration between Transcell Biolife and InARI will create one of the unique biobanking facilities with integrated intelligent systems for disease risk prediction and diagnostics.

This facility is available for only Transcell Biolife’s Karnataka clients.