Science Compilation on Covid-19
One Page of Peer Reviewed Publications, Fighting Against Misinformation.
If you are a scientist and have verified information on this page, or see errors to correct, please message me. The goal is to consolidate all the correct data so we can work together.
(Basics of Covid-19 at the bottom of this page. Covid-19 is the name of the disease. SARS-CoV2 is the virus.)
Grants, Prizes, Submissions:
- NIH’s Urgent Award Grant: If you are working on “viral natural history, pathogenicity, transmission” or “animal models for pre-clinical testing of vaccines and therapeutics against 2019-nCoV”, please do apply.
- Ginkgo Bioworks committing $25 million of free access to its platform for Covid-19 projects
- BARDA submission of your technology: you must own full IP rights or freedom to operate. Categories are vaccine, diagnostic, therapeutic, other. BARDA = US Health and Human Services’ Biomedical Advanced Research and Development Authority.
- Covid-19 Therapeutics Accelerator funded by the Gates Foundation, Wellcome, and Mastercard. “Drugs or monoclonal antibodies that pass initial screening would then be developed by an industry partner.”
- George Mason Univ’s Emergent Ventures has several prizes on treatment, social tracking, policy and journalism associated with Covid-19
- List of projects and funding by the Druid Initiative, SolveCovid19.net
Complete Viral Sequence: list of sites to download the complete viral sequence (due to mutations, the sequence may not stay the same)
Spike protein (S Protein):
- Five German institutions partnered together, and showed that SARS-CoV-2 uses the SARS55 CoV receptor, ACE2, for entry, and the serine protease TMPRSS2 for S protein priming. Their publication in Cell (PDF format) is slow to load.
- Feb 19, 2020 Univ of Washington shows SARS-CoV2 spike (S) glycoprotein trimers mediate entry in VeroE6 cells and in BHK cells transiently transfected with human ACE2. And the differences btw the spike protein of 2003 CoV and 2019 CoV2.
- Feb 1, 2020 publication showing the binding site of the spike protein and human ACE2 receptor.
- Spike glycoprotein of the 2003 SARS virus is very similar (source: UniProt).
- June 2019 Nature article on the S glycoprotein of the family of coronaviruses, engaging 9-O-acetyl-sialogycans in order to gain cell entry.
- February 2019 Cell article noting that 2003 SARS-CoV S glycoprotein antibody triggered fusogenic conformational changes via receptor functional mimicry. The activation of coronavirus membrane fusion takes place through a receptor-driven ratcheting mechanism.
- SARS-CoV-2 invades host cells via CD147-spike protein (March 14, 2020, not peer reviewed)
- One caveat on Spike protein vaccines: Feb 21, 2019 (last year) publication in JCI Insight (American Society for Clinical Investigation) showed that anti–spike IgG causes severe acute lung injury by skewing macrophage responses.
Receptor Binding Domain (RBD)
- Cellular and Molecular Immunology March 19, 2020 development of RBD protein as a viral attachment inhibitor and vaccine
Main viral protease: contact Martin Walsh at the UK Diamond Light Source synchrotron. Post from the Diamond synchrotron is here, and the crystal structure of COVID-19 main protease in complex with an inhibitor N3 is here.
Cultured virus by Sunnybrook, McMaster University and the University of Toronto (March 12, 2020)
ATCC.org’s coronavirus resource page
World Health Organization’s report, PDF format, dated Feb 16–24
Wuhan Center for Disease Control — Coronavirus Prevention Handbook, 101 Science-based Tips (not peer reviewed, Dropbox PDF)
- American Chemical Society March 6, 2020, vaccines and therapeutics
- IBCC Internet Book of Critical Care (not peer reviewed, authored by the attending intensivist at the University of Vermont)
- Lancet Feb 22, 2020, “cytokine storm, which is characterised by increased plasma concentrations of interleukins 2, 7, and 10, granulocyte-colony stimulating factor, interferon-γ-inducible protein 10, monocyte chemoattractant protein 1, macrophage inflammatory protein 1 alpha, and tumour necrosis factor α.”
- JAMA April 13, 2020, overview of pharmacologic treatments for Covid-19
Ways to fight Covid-19: (many places we can intervene before the virus replicates sufficiently to cause damage: at the binding site (Spike S protein), immune response via vaccines and antibodies, or slow/stop viral replication)
- Traditional vaccines — attenuated live viruses. Pros: very effective. Cons: most expensive manufacturing process, long FDA approval (possible fast track given the pandemic).
- Milken Institute’s Covid-19 vaccine tracker(click on “view Covid-19 tracker” at bottom of the page, opens a PDF file)
- Sanofi had been working on a conoravirus vaccine after the 2003 SARS, but antigen manufacturing has been the bottleneck. Sanofi is exploring replacing chicken egg with insect cells for antigen manufacturing. Sanofi uses recombinant DNA, with capabilities boosted by its 2017 acquisition of Protein Sciences for $750 million. Phase 1 clinical trial planned between March 2021 and August 2021.
- GSK continues to work on its adjuvant platform to strengthen vaccine response, working with Chinese biotech Clover Biopharmaceuticals on COVID-19 S-Trimer, currently preclinical. GSK is also working with CEPI (Coalition for Epidemic Preparedness Innovations, a global foundation) grantee University of Queensland.
- UT Austin vaccine delivery via film — if effective, no needles, thermal stable.
- Univ of Pittsburgh’s microneedle array thermostable patch vaccine, shows antibody peaking after 2 weeks in mice
- Jackson Labs announced on Feb 24, 2020 that they have transgenic mice with humanized ACE2 receptors for Covid-19 vaccine testing
- CanSino’s Ad5-nCoV uses its adenovirus-based viral vector vaccine platform. In 2017, the same technology helped the CanSino earn a Chinese approval for its Ebola vaccine, the first Ebola shot approved. Phase 1 trial will be conducted in healthy adults 18 to 60 years of age in Wuhan.
- The bacillus Calmette-Guérin (BCG) vaccine (not peer reviewed) is an existing vaccine for tuberculosis. The Netherlands is looking to give the BCG vaccine to its healthcare workers.
2. Speeding up vaccine production — GMP (good manufacturing practice) of vaccines can take 12 months. Synthetic biology could speed up this process. Ångström Lab at Uppsalla University in Sweden has developed efficient viral removal bioprocessing for CHO cells. CalPoly scientists published a cell-free protein synthesis protocol. Overview of cell-free protein synthesis research.
3. mRNA vaccines — between DNA and protein, there is a middle step carried out via messenger RNA (mRNA).
- The vaccine Moderna submitted to the FDA on Feb 24th uses mRNA. First small human trial underway in Washington State (45 healthy volunteers between the ages of 18 and 55. Start date March 19, 2020. Estimated completion date June 1, 2021). The ClinicalTrial.gov site for the Moderna mRNA study.
- Acturus Therapeutics and Duke Med School developing a mRNA vaccine for Singapore: “The STARR™ Technology platform combines self-replicating RNA with LUNAR®, a leading nanoparticle non-viral delivery system, to produce proteins inside the human body.” “Superior immune response and sustained protein expression … expected to produce a vaccine response at much lower doses compared to traditional mRNA vaccines.”
4. Peptide vaccines
- Inovio Pharma, another CEPI grantee, is working on a peptide vaccine, INO-4800, a synthetic genome and does not contain any of the virus. Clinical trials in the U.S. with 30 participants started April 7th. It also plans to launch human trials in China and South Korea in April, with a total of 3,000 doses.
- Andre Watson of Ligandal seeking funding
5. Viral-vector vaccines — CanSino Biologics is co-developing its viral-vector vaccine (Ad5-nCoV) with the PLA’s Academy of Military Medical Sciences, received approval from China to begin clinical trials on 108 healthy adults in April in Wuhan, China. Trial estimated to conclude at the end of the year.
5. Monoclonal antibodies — instead of a waiting 6–9 weeks for antibodies to build after the vaccine, deliver the antibodies directly. However, half life of delivered antibodies is several days, versus months or years for immune response generated antibodies.
- Jacob Glanville of Distributed Bio is testing antibodies from the 2003 SARS for use against Covid-19 in vitro.
- This Korean study identified CR3022 neutralizing antibody from humans as having high binding affinity with SARS-CoV-2 S protein.
- Regeneron announced on Feb 4 that it’s using its VelocImmune platform (genetically-engineered mice with humanized immune systems) to make prophylactic doses ready for human testing by end of August.
- Vir Bio announced on Feb. 25 it is collaborating with Shanghai-based WuXi Biologics to test monoclonal antibodies as treatment, with Biogen for production
- This Dutch group found an antibody and is trying to find a production partner while getting regulatory approval.
6. Small molecules — mini protein binders are stable even at high temperatures (ease of transport and delivery), but the small size of the molecule, while cheaper to produce, doesn’t often elicit enough of an immune response. Computer designed to adhere only to specific grooves on the SARS-CoV-2 spike protein.
7. Plasma infusion — China has already been trying plasma infusion (dated Feb 17, 2020, not peer reviewed) from recovered patients. Takeda said March 4 it plans to test hyperimmune globulins (called TAK-888) for people who are at high risk for Covid-19. Another term is antisera/antiserum from recovered patients.
8. Antivirals — because scientists have been working on HIV for decades, we have several antiviral drugs that have been tried on Covid-19. The advantage of these antivirals is that many had already started or gotten FDA approval for another disease. (World Health Organization’s PDF list of therapeutics, use and dose for each, and clinical stage)
- Nucleotide blocker — Remdesivir, made by Gilead, misshapen version of a nucleotide that the virus needs to build new copies of itself, thus preventing it from multiplying. Works on RNA viruses, which coronaviruses are. Had been used to treat Hepatitis C. Showed some efficacy in treating MERS, but not Ebola. On March 3, 2020, Gilead announced a 600-patient phase 3 trial of Remdesivir for the treatment of Covid-19. A phase 2 multi-center trial against placebo. Another shorter phase 2 trial. Early trials in China showing better results with Remdesivir than Kaletra.
- Protease inhibitor — in order for viruses to enter the cell and replicate itself, it uses enzymes to dissolve cell proteins. These enzymes are called proteases. Protease inhibitors are anti-proteases. Lopinavir and Ritonavir, made by AbbVie as a combo drug Kaletra in the U.S., was tried on 52 patients in China (link in Chinese), without much effect. March 18, 2020 NEJM publication states that Kaletra was no better than standard care for 199 patients. German study on camostat mesilate (March6, 2020, not peer reviewed) showed efficacy in inhibiting the protease TMPRSS2, on viruses isolated from a SARS-CoV2 patient. The PDF of the publication in Cell is slow to load. Camostat mesilate is approved in Japan for use in pancreatic inflammation.
- Increasing endosomal pH — a certain pH level is necessary for viral-cell fusion. Raising this pH level slows or stops viral-cell fusion. Chloroquine is a drug to treat malaria, sold under Aralen in the U.S. Chloroquine may also interfering with the glycosylation of cellular receptors of SARS-CoV. Chloroquine could prevent ORF1ab, ORF3a and ORF10 to attack the heme to form the porphyrin, and inhibit the binding of ORF8 and surface glycoproteins to porphyrins. Sometimes mefloquine or atovaquone are used as replacements. China has already tried an in vitro study of Chloroquine and Remdesivir (Nature, Feb 4, 2020). * Clinical Infectious Diseases March 9, 2020 (not peer reviewed) publication, hydroxychloroquine is potentially more potent than chloroquine (in vitro). * Nature March 18, 2020 article shows hydroxychloroquine is less toxis than chloroquin in vitro. * University of Minnesota is running a post-exposure (≤ 3 days) prophylaxis trial of hydroxychloroquine with 45 patients in 19 states, firstname.lastname@example.org for info. * French study March 28, 2020 showing no evidence of clearance or clinical benefit for Hydroxychloroquine +Azithromycin in severe Covid-19 (not a randomized trial) (in English)
- Clinical Infectious Diseases March 9, 2020 (not peer reviewed) publication, hydroxychloroquine is potentially more potent than chloroquine (in vitro). This Nature March 18, 2020 article shows hydroxychloroquine is less toxis than chloroquin in vitro. University of Minnesota is running a post-exposure (≤ 3 days) prophylaxis trial of hydroxychloroquine with 45 patients in 19 states, email@example.com for info.
- Adenosine nucleotide analogue — an adenosine nucleotide analog that confuses viral RNA polymerase and evades proofreading by viral exoribonuclease (ExoN), causing a decrease in viral RNA production. Remdesivir, made by Gilead, metabolizes into its active form GS-441524, which is a nucleotide analogue. It was created to treat single RNA viruses such as Ebola and MERS.
- Make viruses unable to detach after replication — Nitazoxanide, usually a antiparasitic, entered phase 3 clinical trials for use as a neuraminidase inhibitors antiviral. Neuraminidase inhibitors (NAIs) block the neuraminidase enzyme: it blocks the budding off of viruses from the host cell after replication. Neuraminidases are surface enzymes on influenza viruses that cleave sialic acid groups from glycoproteins, which buds off the new virus from the host cell. All human H#N# viruses have neuraminidase enzyme, where the H refers hemagglutinin and N refers to neuraminidase. Oseltamivir (Tamiflu), Zanamivir (Relenza), Laninamivir (Inavir), and Peramivir, are all NAIs.
- Inhibit virus-cell membrane fusion — Umifenovir, sold under Arbidol in Russia and 阿比朵尔 in China has been used in China (in Chinese) to treat Covid-19. Umifenovir prevents contact between the virus and target host cells, and inhibits fusion between the viral envelope (surrounding the viral capsid) and the cell membrane. Japan’s experience using Arbidol in combination with other drugs (in English).
- Nucleotide reverse transcriptase inhibitor (RTI) — RTIs inhibit activity of reverse transcriptase, a viral DNA polymerase that is required for replication of retroviruses (RNA viruses that replicate via reverse transcriptase, and inserts viral genome into the host DNA via integrase enzyme). Tenofovir alafenamide (TAF) and emtricitabine, marketed as Descovy for HIV PrEP by Gilead. Had been used to treat HIV and Hepatitis B.
9. Antivirals NOT showing high efficacy:
Antivirals NOT showing high efficacy:
- Confuse ‘em— Ribavirin is a nucleoside inhibitor, specifically a guanosine ribonucleic analog, by confusing RNA metabolism required for viral replication. It metabolizes into a molecule resembling purine RNA nucleotides. Ribavirin unfortunately has many side effects.
- Penciclovir, also a nucleoside analogue, used to treat herpes. Sold under Denavir, Vectavir, Fenivir, Famciclovir. Within days in the human body, penciclovir adds three phosphate groups, producing the active penciclovir triphosphate, inhibits viral DNA polymerase, thus impairing the ability of the virus to replicate. Selectivity of penciclovir means negligible cytotoxicity to healthy cells.
- Induce mutations — Favipiravir, aka T-705, Avigan, Favilavir, made by Toyama Chemical (Fujifilm group) in Japan, inhibits viral RNA polymerase and induces RNA transversion mutations such that the viral phenotype is no longer viable. However, it has limited effectiveness in primary human airway cells.
- Nafamostat, serine protease inhibitor. Usually used as an anticoagulant during hemodialysis to prevent the proteolysis of fibrinogen into fibrin.
Antivirals not yet tried (haven’t seen research on it):
- Aciclovir — nucleoside analogue very similar to penciclovir, used to treat cytomegalovirus such as herpes, Epstein-Barr, chicken pox, and shingles.
- Mefloquine or atovaquone — Chloroquine similars
- This accepted but not peer reviewed article cites neuroinvasive potential of SARS-CoV2 as partially responsible for respiratory failure
10. Interferons — I’ve seen interferon-1b mentioned as having antiviral functions, but this study from the National Cancer Institute in 2014 on antiviral immunotherapies cautions against its side effects. This Feb 2019 publication in Viruses on the difference between bat immune system and other mammals hints at why might interferons work.
11. Janus Kinase (JAK) inhibitors — This Feb 4, 2020, Lancet publication used Benevolent.ai to cull through structured medical data and identified JAK inhibitor baricitinib as a candidate. Baricitinib is currently made by InCyte and Lilly under name Olumiant to treat rheumatoid arthritis. Side effects include upper respiratory infections.
12. Other Immune Responses: (anti-inflammatories are only recommended in the second phase of the infection, the ARDS phase. Not the first phase.)
- Arthritis drug Sarilumab blocks interleukin 6 (I-6). Sanofi and Regeneron announced on March 16 a two-phase trial of sarilumab on patients with severe cases of COVID-19 in New York. Eventually will include 400 patients at 16 U.S. sites.
- Roche starting phase III trials for Actemra/RoActemra (tocilizumab). The drug is originally made for rheumatoid arthritis and targets IL-6.
- Nature March 16, 2020 immune responses prior to patient recovery: a case report of non-severe COVID-19. “Increased antibody-secreting cells (ASCs), follicular helper T cells (TFH cells), activated CD4+ T cells and CD8+ T cells and immunoglobulin M (IgM) and IgG antibodies that bound the COVID-19-causing coronavirus SARS-CoV-2 were detected in blood before symptomatic recovery.” “In contrast to severe avian H7N9 disease, which had elevated cytokines IL-6, IL-8, IL-10, MIP-1β and IFN-γ6, minimal pro-inflammatory cytokines and chemokines were found in this patient with COVID-19, even while she was symptomatic.”
- Journal of Emerging Microbes and Infections March 20, 2020 suggests “IVIg and low molecular weight heparin (LMWH) anticoagulant therapy”.
12. Chinese traditional medicine — [Please note this is meant to be in combination with antivirals, not in place of. AND, not OR.] A combo herbal drink has already been administered in southern province of Guangdong, and coastal Zhejiang with 0.1% and 0% mortality rate respectively. This is the one published article I could find on the possible scientific explanation of why: Oct 2014 Nature, Honeysuckle-encoded atypical microRNA2911 directly targets influenza A viruses. The microRNA2911 remain almost intact after being boiled, and tended to concentrate in the mice lungs, doubling survival rate. Currently a 400 participant clinical trial is happening in China on shuanghuanglian, an herb containing extracts from the dried fruit lianqiao (Forsythiae fructus), purported to have been used for treating infections.
13. Glucocorticoids — Circlesonide (branded under Alvesco) (not peer reviewed, translated from Japanese) was used in the treatment of several cases in Japan with various results.
14. Treating ARDS — Feb 2014 publication in Critical Care, overview of pharmacological treatments of ARDS, including nitric oxide inhalation, neuromuscular blocking agents, corticosteroids, angiotensin-converting enzyme inhibitors, statins and stem cells.
15. Bronchodilation? [NOT yet tested for Covid-19] 1982 article from British Journal of Clinical Pharmacology
Viral RNA Persistence — This March 11th Lancet publication notes “detectable SARS-CoV-2 RNA persisted for a median of 20 days in survivors”
ACE2 receptor as the likely method of viral entry: Acute Hemorrhagic Necrotizing Encephalopathy: CT and MRI scans.
- Steps to make your own (Mammoth Bio white paper)
- Rutgers receives emergency FDA approval for its saliva test
- Genentech received an Emergency Use Authorization approval from the FDA and IVD CE-Mark for a Roche Diagnostics for SARS-CoV2 on March 12, 2020
- Mammoth Bio and Sherlock Bio both working on using CRISPR for testing. RNA fragments cut by Cas13 would trigger a band on an at-home test kit just like a pregnancy test. Nature Feb 2019 article on mechanism.
- Oxford University’s RNA and RNA fragment test takes only 30 min
- Serology — Biomerica’s $10 antibody test, finger prick blood test with results in 10 minutes.
- FDA issued emergency authorizations for high-speed test kits to diagnose the coronavirus, including Waltham-based Thermo Fisher Scientific and Marlborough-based Hologic.
- Abbott’s 5-minute point-of-care test (runs on Abbott’s ID NOW platform)
- Nature article Feb 15, 2020 on trials planned by the World Health Org
- Chinese Clinical Trial Registry — there are 200+ clinical trials on Covid-19 in China right now, most are double blind
Patient Data Sets:
- Github repository of chest x-rays
- South Korea Kaggle repository
- Github repository from Johns Hopkins
- Digestive symptoms
FREE Science Journals — Covid-19 publication lists:
- NEJM (New England Journal of Medicine)
- MedRxiv (some not peer reviewed)
- ACS (American Chemical Society)
- JAMA March 4, 2020 study suggests transmission via respiratory droplets and fecal shedding
- JAMA March 26, 2020 turbulent flow suggests 8+ meters of droplet travel
- Nature March 13, 2020 prolonged fecal shedding in pediatric patients
- NEJM March 17, 2020 aerosol and surface stability of HCoV-19 (SARS-CoV-2) compared to SARS-CoV-1. Aerosol behavior slide deck on CDC’s website.
- TedMed talk by the Director of Fluid Dynamics of Disease Transmission Lab at MIT, on SARS-CoV-1 potential aerosol cloud transmission
- Susceptibility and transmission in pets: cats, dogs, ferrets. (not peer reviewed)
- Nature April 1, 2020, susceptibility in cats, dogs, pigs, chickens, ducks.
Track Viral Mutation:
- NextStrain phylogenic tree (each dot shows location and date)
Track Geographic Spread:
- Johns Hopkins map and worldwide table
- log growth vs linear growth by country and lagged days from infection
Exponential Growth vs Logistic Growth:
- video explanation
- growth curves by state: do nothing, vs shelter-in-place, vs social distancing, vs Wuhan shutdown
Artificial Intelligence and Software Tools:
- DeepMind’s AlphaFold is coming up with protein predictions of Covid-19
- Fold.it — video game version of protein folding. Design proteins that can stick to the Covid-19 spike protein.
- Folding@home — protein folding desktop app developed by Stanford, donate some of your unused CPU while it runs in the background
- Rosetta Commons makes beautiful 3D protein folding images. Free download, but takes a few hours to compiled.
- Insilico Medicine in South Korea designed a 3C-like protease inhibitor (in Korean, not peer reviewed). Algorithm takes in three input data types: crystal structure, homology model, and co-crystal ligand.
- Benevolent.ai searches through structured medical info. This UK team identified baricitinib as a potential treatment in this Feb 4, 2020 Lancet publication. Baricitinib is a janus kinase (JAK) inhibitor used to treat rheumatoid arthritis, sold under Olumiant by Incyte and Eli Lilly. Side effects include upper respiratory infections.
- https://covidbase.com/ list of Covid-19 projects
- Kaggle CORD-19 dataset
- Neural network out of MIT trained exclusively on global #COVID19 data (other models included SARS1 MERS) correctly predicted April plateau, but warns “relaxing quarantine measures too soon, we have predicted that the consequences would be far more catastrophic.” (not peer reviewed)
Background on Covid-19: Coronavirus is the name for a family of viruses, including the 2003 SARS, the 2012 MERS. Covid-19 has a lot of protein structural and genomic similarities to SARS: in fact, the scientific name for Covid-19 is SARS-CoV2 (the 2003 SARS is called SARS-CoV). Contrary to popular comments, Covid-19 is NOT “just a flu.” Viral causes of the common cold and flu comprise of 10–40% rhinovirus, ~20% coronavirus, ~20% respiratory syncytial virus (RSV) and parainfluenza virus (PIV). There are over 30 types of coronaviruses that we know of. Although Covid-19 has much lower mortality rate than SARS, its long incubation (and asymptomatic transmission) period is causing it to have a much higher R0. Even if the mortality rate is lower, if the R0 is much higher, the absolute number of deaths could be bigger. Mortality rate in the US is skewed to the high side due to lack of testing, but South Korea offers drive-up testing and its 0.6% case mortality rate (dated March 6) could be much more accurate. This compares to the 0.1% mortality rate of the seasonal flu in the U.S.