● Scalability. ● On-premise hardware investment (capital expenditure) versus cloud-based implementation (operational expenditure). ● Workforce – bioinformaticians, software engi- neers, DevOps teams. ● Compute and storage costs. ● Direct instrument integration. ● Security and compliance requirements, especial- ly in a clinical environment. ● Accuracy and reproducibility. ● Turnaround times. ● Out-of-the-box, plug-and-play solutions versus custom-built.

Integrating the data siloes – analysing separate data sets together Sharing healthcare genomics data across multina- tional sites creates a range of challenges. Interesting work to address some of these is being carried out by Elixir and the Genomic Alliance for Global Health (GA4GH). It has been estimated that in 2012, roughly 1%of all genome sequencing was funded by healthcare; by 2022 that is expected to increase to 80%. As clinical data requires addi- tional security and compliance requirements over research data, most clinical data sets need to remain in defined geographical

for biomarker discovery. The data generated can also be integrated with clinical outcomes data to perform patient survival analyses. Currently ClaraT covers three of the 10 biologies identified to have a role in cancer, with plans to ultimately cover all 10 areas. However, this assay is designated as RUO (Research Use Only) and as such cannot be used for diagnostic or prognostic purposes, including pre- dicting responsiveness to a particular therapy. Machine Learning (ML) has been deployed on

sequencing data derived from FFPE (Formalin Fixed, Paraffin Embedded) tissues to automate removal of artefacts introduced by the formalin- fixing process. This approach has reduced the inci- dence of artefacts from 42%down to just over 1% in some cases. One of the key advantages in doing this is to allow archival samples to be screened more accurately, allowing the possibility of increas- ing the numbers of available clinical trial partici- pants.


However, to benefit from these large volumes of data distributed globally, especially when looking at rare diseases, there needs to be a means to pro- tect patient confidentiality while allowing researchers to search these data sets to identify where specific patient populations can be found. Federating databases is the best practice identified to date, but it requires common data models and tools to allow interoperability across sites. This becomes even more pressing with initiatives

being announced such as the MEGA (Million European Genomes Available) Initiative15. Elixir is working closely with other national organisations and global initiatives, such as GA4GH to address some of these issues with projects such as the Beacons initiative, which allows users to identify the locations of patients with specific genetic char- acteristics across multiple sites globally, and creat- ing a Tools Platform16 to provide easy access to bioinformatics tools.

NGS bioinformatics: challenges and solutions For more diagnostics to be developed to identify more diseases, more biomarkers need to be discov- ered. The ClaraT assay fromAlmac can analyse gene expression data from microarrays or RNAseq data


Emerging technologies – Blockchain andAI/ML Looking forward, Blockchain and Artificial Intelligence/Machine Learning (AI/ML) could play an increasingly important role. Blockchain utilises distributed ledger technology and provides an irre- vocable audit trail for all data handling without the requirement of a trusted party. Such an approach could contribute strongly to making sequence data available to researchers in a con- trolled manner, the partnership between Shivom and Lifebit providing one example of such capabil- ity17. Furthermore, the application of blockchain technology could support the regulatory compli- ance of diagnostic analyses. It was noted that the Pistoia Alliance had a blockchain project ‘Blockchain supporting Life Science&Health’18 to explore the capabilities of this exciting technology in life sciences. It was anticipated that if AI were optimally

exploited it could make a strong contribution to biopharma and healthcare. Some examples were put forward, ie AI could:

● Predict patient drug response. ● Support patient stratification to optimise clinical trials or to personalise treatments. ● Predict disease progression. ● Diagnose disease. ● Discover biomarkers (thereby improving diag- nostics). ● Optimise drug design in silico to increase effica- cy and decrease toxicity. ● Support multi-omics data analysis optimisation.

Drug DiscoveryWorld Summer 2019

Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64