MedGenome revolutionizing Genomic Sequencing Technologies with Bioinformatics
CXOToday has engaged in an exclusive interview with Dr Ravi Gupta, Vice President, Bioinformatics, MedGenome
- How has bioinformatics revolutionized genomic sequencing technologies, making them more accessible and affordable for widespread diagnostic applications?
Bioinformatics field has played a pivotal role in revolutionizing the field of genomics. It has accelerated progress and amplified the success of genomic sequencing technologies and their applications. For e.g., the human genome sequencing which took more than a decade to assemble the first human draft genome; now analysis can be done in less than an hour. Furthermore, the cost of analysis has come down to less than 100 dollars as compared to millions spent earlier.
Faster genomic data analysis has opened up several applications including newborn screening for which analysis and interpretation of the human genome can be done in a few hours. It equips clinicians with the correct information as early diagnosis of genetic disorders helps in improving the health of the child and in some cases reduces the mortality rate.
Advancement in bioinformatics algorithms has also led to faster and more accurate clinical data interpretation. Variant interpretation from genomic data is a humongous task and takes several hours to identify the disease-causing variants. With the advancements in bioinformatics software, the variant interpretation can be accurately done in less than an hour for most cases. The latest bioinformatics algorithms are much faster and less resource intensive which enables faster and cheaper analysis and interpretation of the clinical diagnostics data. All these advancements have led to large scale adoption of genomic sequencing technologies which has led to reduction in cost and many broader applications.
- Can you elaborate on the role of artificial intelligence and machine learning in bioinformatics-driven genomic analysis, and how it is enhancing the accuracy and efficiency of diagnostic processes?
Machine learning and AI is now applied at different stages of genomics data analysis. For example, the deep neural network learning based variant caller such as DeepVariant from Google is helping in accurate variant calling. DeepVariant was also recently applied in tandem with oxford nanopore for newborn screening application. The variant caller by Illumina in DRAGEN and Sentieon variant callers are now trained and use machine learning to improve variant calling from next-generation sequencing dataset. Solutions such as VaRTK from MedGenome, Invitae’s Clinical Variant Modeling, Illumina’s Emedgene are trained on various clinical data and use various machine learning techniques to accurately identify pathogenic variants from millions of variants. Such applications/solutions have made clinical data interpretation much faster and more accurate. Google DeepMind group recently released AlphaMissense, a deep learning model trained on large protein structure prediction tool AlphaFold2. Illumina published a deep learning tool SpliceAI which helps in identifying damaging splice variants. Researchers are also making attempts to implement Large Language Models (LLMs) for genomic interpretation. Overall, artificial intelligence and machine learning along with bioinformatics tools are driving analysis and interpretation of clinical diagnostics dataset and enabling faster and accurate diagnostics.
- Considering the rapid advancements in bioinformatics, what future innovations do you foresee that could further accelerate the development and adoption of genomic sequencing for precision medicine?
Generating faster and accurate full proof clinical level reports from raw sequenced data is one of the key challenges that the industry is facing. Although there are several solutions available, a clinical grade fully automated report software is something that the genomics driven diagnostic field is looking for. To develop such software that can generate reports from complicated data analysis needs future innovation to further accelerate and scale up the adoption of genomics sequencing for precision medicine. This will help in making advanced sequencing-based techniques to become like a regular diagnostic tool for the clinician that can benefit all types of patients – from newborn, early cancer detection to complicated diseases.
