Precision Medicine Approaches to Ending the Coronavirus Pandemic

Last updated: 12 March 2021

This article addresses the following key questions:

1. What is biological intelligence (BI) and how can it help us reach COVID-Zero?
2. What are Pharma 1.0 drug discovery approaches suggesting against COVID19?
3. What are Pharma 2.0 AI-driven drug discovery suggesting against COVID19?
4. What are Pharma 3.0 BI-driven drug discovery approaches suggesting against COVID19?
5. How does BI leverage Genomics England Biomarker Panel Data to enable Precision Medicine for COVID-19 infected patients?
6. What efforts may help us deliver COVID19 treatments?

Q1: What is biological intelligence and how can it help us reach COVID-Zero??

Biological intelligence is a biology first approach that enables failure free drug discovery without deep learning steps associated with artificial intelligence (AI). you have 4 minutes you may be interested to read https://lnkd.in/daS8EQk that is Danny Sullivan's feature on BRAINCURES at Longevity.Technology from 4 June 2020 entitled "Move over AI? Here comes biological intelligence"

As shown above, BI is biology-powered approach that harnesses the best out of the traditional and artificial intelligence (AI)-driven drug discovery approaches of today to shorten the drug discovery cycle by at least 4 years and predict compounds that have an at least 6-fold higher success rate compared to today's 1 in 10 success rate from Phase I to approved medicine. In this case study we will apply our in silico biology first approach to (i) identify and test approved drugs for activity against COVID-19; (ii) screen libraries of thousands of compounds with confirmed safety data and (iii) consider new investigational compounds and monoclonal antibodies (see lnkd.in/gbk2gda for more details).

Of note BRAINCURES' approach has allowed us to find that (i) 5 -(Interferon Beta-1a, Interferon Beta-1b, Ruxolitinib, Baracitinib, Canakinumab) of 61 BRAINCURES prioritized COVID19-Drugs are in Clinical Trials; (ii) 4 further drugs have relevant clinical evidence generated prior to the coronavirus pandemic; and (iii) 48 further drugs have relevant preclinical evidence.

Q2. What are Pharma 1.0 drug discovery approaches suggesting against COVID19?

The following table lists some of the current drug repositioning efforts discussed within Nature dated 27 February 2020 (see lnkd.in/gjA4v-T)

Q3: What are Pharma 2.0 AI-driven drug discovery approaches suggesting?

There are nearly 200 startups that employee AI to some aspect of the precision medicine life cycle (e.g., compound generation, target lead selection, biomarker prediction, virtual phenotypic screening, virtual patient matching in order of most often to least often offered). Some notable outcomes of Pharma 2.0 efforts include:

• 1005 unique drugs and 12053 unique genes have been linked to COVID-19 within the Crowd Generated Gene and Drug Set Library Resource that comprises 62 drug sets and 306 gene sets (see https://amp.pharm.mssm.edu/covid19/ last accessed 25 April 2020). One noteworthy study, applied a network pharmacology approach to prioritize 16 of 2938 evaluated drugs:

Source: https://www.nature.com/articles/s41421-020-0153-3

• the quite rapid by industry standards generation of 6 compounds within fewer than 25 days. While some media reports have incorrectly dubbed this as drug repositioning, the results of this study are limited as the authors have not performed any preclinical efficacy testing of the predicted compounds. Thus, the clinical use of these predicted compounds will be very limited in the war against COVID19 today if the traditional and safe for human use drug development life cycle is to be followed (see lnkd.in/gQNy7gf) until in vitro and in vivo preclinical data is obtained:

• the nomination of a cannabis-like compound that might threat superbug resistance (see lnkd.in/gMRmjnq). Structures and the phenotypic effect of a selected compound that was screened is shown below:

• 20 articles on how scientists are using AI and data science to fight COVID19

Thus there still remains a largely uncharted opportunity to apply data science approaches to identify single existing drugs or combinations of thereof for the treatment of patients infected with antimicrobial resistant bugs and/or COVID19 across the world.

Q4: What are Pharma 3.0 BI-driven drug discovery approaches suggesting against COVID19?

The BDE algorithms and processes prioritized 31 of 266 Coronavirus linked target genes. These BDE targets are supported by 59 coronavirus relevant publication. Further deeper analysis, reveals that 14 of the 31 prioritized targets are modulated by 107 Phase IV ready drugs.

Please contact us to discuss the de-anonymization of the remaining 9 non-immune-related BDE Targets or the ranking of these drugs based on text-mining prioritization or virtual L1000-BDE double phenotypic screening.

Q5: How does BI leverage Genomics England Biomarker Panel Data to enable Precision Medicine for COVID-19 infected patients?

The BDE can accelerate biomarker-based patient selection by prioritizing 15 of the 508 viral susceptibility-associated genes contained within the Genomics England (GeL) PanelApp for use at mass scale.

In more detail:

• we first identified the viral susceptibility panel from 324 panel catalogued within the Genomics England PanelApp Resource at https://panelapp.genomicsengland.co.uk/. This panel contains 508 (420 green, 17 amber, 71 red) biomarker genes. The 420 Green lighted genes on this panel have the highest clinical readiness, while the amber- and red- lighted genes contain borderline or low levels of evidence that limit their utility in the clinic
• we then applied the BDE molecular framework of 822 targets to prioritize 33 (24 green, 1 amber, 8 red) viral susceptibility targets
• finally we recommend 15 of the 33 BDE-prioritized viral susceptibility genes to be used for patient stratification at scale (contact us to de-anonymize the finding)

Q6: What efforts may help us deliver COVID19 treatments?

Q7: Other coronavirus relevant info

and

and

Thank you for your attention and do come back regularly to check for updates on future analysis results that we will be posting here each week.