Viruses and Phages

"there is a lack of in-depth analysis and theory building in the creationist literature regarding viral function" — Joseph Francis, Matthew Ingle and Todd Wood, International Conference on Creationism, 2018¹

"viruses kill about 20% of the living material in the ocean every day" — Curtis Suttle, 2013²

Summary

Young-earth writers say viruses were created good and that the ones causing disease are corrupted, displaced, or escaped from the genomes they once served.  Part of that is measured.  Part of it is contradicted.  The rest has not been tested by anyone, including its authors.

• The young-earth position, in its own words. Joseph Francis wrote in 2003 that "all microbes and viruses were created as mutual symbionts" and that pathology arose "probably after man's Fall." Jerry Bergman, Todd Wood, Yingguang Liu and Peter Borger add that harmful viruses are mutated or misplaced versions of useful gene carriers, and that some viruses began inside host genomes. In 2018 Francis, Ingle and Wood said plainly that the creationist literature on what viruses do is thin.³¹
• Most viruses are not making anyone sick. About 10^{fuhrman-1999:mortality][} most of them infecting bacteria. Ocean viruses carry out about 10^{fuhrman-1999:mortality][}d. Measurements put viruses behind 10 to 50 percent of bacterial deaths in surface seawater. A healthy person carries about 10^{fuhrman-1999:mortality][}man-1999:mortality]⁴⁵⁶
• Regulation is real, and it cuts both ways. In Dhaka, cholera epidemics tend to begin when cholera phage is scarce and end as it becomes abundant. But the cholera toxin itself is carried by a different phage. Cyanophages carry working photosynthesis genes, and use them to keep a dying cell productive for the virus.⁷⁸⁹
• The pathogens with dated origins arrived by host jump or gene capture. The Rous sarcoma virus cancer gene is a chicken gene with its off switch replaced. HIV-1 entered humans from chimpanzees around 1900. Measles split from a cattle virus within recorded history. Myxoma virus gives its native host a skin lump and killed 99.8 percent of European rabbits. But the source host is not always unharmed: the chimpanzee virus kills chimpanzees, and the cattle virus was a cattle plague.^10111213
• The escaped-gene model fails its clearest test. Borger wrote that "most RNA viruses are built of genetic information directly related to that of their hosts." Eugene Koonin's survey found the reverse: small RNA viruses are made mostly of genes "missing from cellular genomes." Liu and Soper concede that humans have no endogenous relative of HIV.¹⁴¹⁵¹⁶
• Verdict. What is measured fits a world in which viruses mostly regulate and occasionally go wrong. What is asserted is that they were all once harmless. The design answer is weakest where a virus carries a finished weapon, such as the cholera toxin, that works well on human cells.

Part 1: What young-earth writers propose

Three claims, from their sources

The literature is small.  Francis, Ingle and Wood listed it in 2018 and then wrote that "there is a lack of in-depth analysis and theory building in the creationist literature regarding viral function."  It is spread across the Journal of Creation, Answers Research Journal, Origins and the International Conference on Creationism.  It makes three separate claims.¹

Viruses were created good.  Francis's 2003 conference paper holds that "all microbes and viruses were created as mutual symbionts—that is, without pathology.  Pathology arose after the creation, probably after man's Fall."  He predicted that disease-causing microbes would be rare and that their weapons would turn out to be "modified symbiotic mechanisms."  Francis, Ingle and Wood repeat in 2018 that viruses "were created as part of God's original good creation."³¹

Harmful viruses are mutated or misplaced.  Bergman wrote in 1999 that "in a post-Fall world, disease is a result of 'something gone wrong'".  He meant a mutation or genes moving where they should not.  Wood's 2002 model of helpful mobile DNA, which he called altruistic genetic elements or AGEs, says mutations "changed the benignly reproducing AGEs into virulent pathogens."  Jean Lightner of Answers in Genesis wrote that "it is quite reasonable to believe that disease-causing viruses are descended from viruses that were once not harmful."  Jonathan Sarfati of Creation Ministries International wrote that "Ebola virus had a benign function before the Fall, as it seems to have today in bats."^17181920

Some viruses escaped from genomes.  Liu and Soper proposed in 2009 that retrovirus-like sequences were built into animal genomes and that free retroviruses came out of them:  "Thus we speculate exogenous retroviruses were derived from endogenous retroviruses."  Borger, writing as Peer Terborg, went further.  He renamed mobile DNA "variation-inducing genetic elements," or VIGEs, and wrote that "RNA viruses have emerged from VIGEs" and that "the RNA viruses got their genes from their hosts."  He later called this the VIGE-first hypothesis.  Brian Thomas of the Institute for Creation Research wrote in 2020 that RNA viruses being harmless in their original hosts "points to their origin within the genome of their host."^1516162122

What each claim has to show

The three claims carry different burdens, as the parent article sets out for every creationist origin model.  The first needs a named benefit and a named beneficiary.  The second needs the harmless ancestor and the change that spoiled it.  The third makes the sharpest prediction:  if viruses came out of host genomes, viral genes should look like host genes.

One point of fairness.  A phage that regulates bacteria does it by killing them.  None of these authors claims otherwise, and none treats the death of a bacterium as the kind of death that Genesis ties to the Fall.  The benefit they claim is to the animal, the plant or the ecosystem.

Part 2: What viruses do when nobody is sick

The numbers

Arcady Mushegian re-derived the standard count in 2020 and found that the number of virus particles on earth is close to 10^{mushegian-2020:close][}either much less or much more than that."  His table puts 2 × 10^{mushegian-2020:close][}er beside 1.2 × 10^{mushegian-2020:close][}.  The figure is a product of estimates, not a census.⁵⁵

Curtis Suttle's 2007 review gives about 10^{fuhrman-1999:mortality][}e ocean each second and calls them "a major source of mortality."  In a 2013 lecture he put it as "viruses kill about 20% of the living material in the ocean every day."  That sentence is a summary.  The measurements behind it are in Jed Fuhrman's 1999 review:  viruses "are responsible for about 10–50% of the total bacterial mortality" in surface waters.  The two popular figures agree with each other only roughly.  At 10^{fuhrman-1999:mortality][} about 7 percent of the ocean's free-living bacteria are infected each day.²³⁴²²⁴

So the regulator role is established by secular oceanographers.  Dead cells release their contents, and other microbes take them up.  Without that turnover the surface ocean would work differently.

Cholera: one phage brakes it, another built it

Shah Faruque's group sampled two rivers and a lake in Dhaka every month from January 2001 to November 2003.  Of 221 water samples, "114 samples (51.5%) contained either a vibriophage or an epidemic V. cholerae, whereas only 15 samples (7.2%) contained both."  In 10 of those 15 the bacteria were resistant to the phage beside them.  Epidemics, they concluded, "would most likely begin in periods of low phage concentration (after floods and the monsoon season)."⁷⁷²⁵

A second study followed one epidemic in 2004 every two to three days for 17 weeks and tested the stools of 387 patients.  Phage built up inside the patients and then in the water, and the epidemic ended.  The authors wrote that phage amplification in patients and phage predation in the environment "may explain the self-limiting nature of seasonal cholera epidemics in Bangladesh."  A 2006 model found that "vibriophage can ameliorate cholera outbreaks."  These are correlations and a model.  Nobody has added phage to a river and stopped an epidemic.²⁶²⁷

Francis, Ingle and Wood cite this work, but describe it as a phage acting on cholera bacteria "in the ocean."  The samples were fresh water in a city.⁷¹

The same research group supplies the other half of the story.  Vibrio cholerae is ordinary water life.  The epidemic strains "evolved from environmental non-pathogenic strains by acquisition of virulence genes," and the toxin genes "are encoded by a lysogenic bacteriophage."  A lysogenic phage is one that inserts itself into the bacterium's DNA and stays.  So one phage limits cholera and another phage makes cholera possible.⁹⁹

Francis and Wood faced this in 2013 and deserve credit for it.  They wrote that toxins which "appear to be created to interact with human cells" are "difficult to explain within the context of a 'good creation.'"  Their answer is that the toxin "was designed to be a modular transmissable factor" for bacteria living as partners inside sea animals, perhaps to manage salt.  They searched for the toxin genes in the squid partner Vibrio fischeri and did not find them.  The proposal remains a proposal.²⁸²⁸²⁸

Cyanophages: photosynthesis genes in a virus

Prochlorococcus and Synechococcus are the most numerous photosynthesizers in the sea.  Matthew Sullivan's survey of 42 of their phages found that "88% of the phage genomes contain psbA," the gene for a core protein of the light-harvesting machinery.  Debbie Lindell showed in 2005 that the phage copy is switched on during infection while the cell's own copies shut down, and that "replication of the phage genome is a function of photosynthesis."  Darkness cut phage DNA copying fourfold.⁸²⁹

The virus keeps the lights on because it needs the power.  The cell still bursts.  The phage genes cluster with the host's in family trees, "indicating that they are of cyanobacterial origin."  That matters for Part 4.  When a virus does take a gene from its host, the match is easy to see.³⁰

The viruses in a healthy person

Guanxiang Liang and Frederic Bushman's 2021 review puts about 10^{moustafa-2017:fortytwo][}h person and about 10^{moustafa-2017:fortytwo][}and notes that "healthy humans are also colonized by a remarkable diversity of viruses."  Most are phages.  A study of blood from 8,240 people, "none of whom were ascertained for any infectious disease," found human virus sequences "in 42% of the study participants."  That list includes hepatitis B and HIV, so it measures silent carriage, not harmlessness.³¹⁶

The gut does not behave like the ocean.  Alejandro Reyes followed twins and their mothers for a year and found that more than 95 percent of each person's virus types stayed put.  The kill-and-replace pattern seen in seawater "is notably absent in the very distal intestine."  Gut phages mostly sit inside bacterial DNA and wait.  Francis, Ingle and Wood's proposal that gut phages guard the microbiome extends the ocean pattern to a place where it has been looked for and not found.³²

Measured benefits

Four cases are solid.

• A virus that lets a plant take heat. A grass growing in hot soil at Yellowstone depends on a fungus, and the fungus depends on a virus. "Fungal isolates cured of the virus are unable to confer heat tolerance, but heat tolerance is restored after the virus is reintroduced." Remove and restore is the strongest kind of evidence in this article.³³
• A virus that stands in for gut bacteria. In mice raised without bacteria, infection with mouse norovirus "restored intestinal morphology and lymphocyte function without inducing overt inflammation and disease."³⁴
• Phages in mucus. Jeremy Barr found phages concentrated in the mucus of animals from corals to humans and proposed that they provide "a ubiquitous, but non–host-derived, immunity." The parent article covers this.³⁵
• Viral parts built into animals. Parasitoid wasps make virus-like particles from genes of an ancient insect virus, and mammals build the placenta with proteins "captured and domesticated" from retroviruses. See Parasitoid Wasps and Shared Retroviruses and Pseudogenes. The wasp case is a benefit to the wasp and a death sentence for the caterpillar.³⁶³⁷

Marilyn Roossinck, who led the Yellowstone work, summarised the field in 2015:  "Viruses are being redefined as more than just pathogens."  Francis predicted in 2003 that this would happen.  The prediction has held up.  It does not show that every virus began this way.³⁸

Part 3: Where the pathogens came from in the measured cases

A cancer gene taken from the chicken

Peyton Rous passed a chicken tumour with a cell-free filtrate in 1911.  In 1976 Dominique Stehelin, Harold Varmus, Michael Bishop and Peter Vogt showed that the virus's cancer gene, src, has a relative in healthy birds:  "the DNA of normal chicken cells contains nucleotide sequences closely related to at least a portion of the transforming gene(s)."  Varmus and Bishop received the 1989 Nobel Prize "for the discovery of the cellular origin of the retroviral oncogenes."³⁹⁴⁰

The captured gene is damaged.  Sequencing showed that "only a portion of c-src is represented within v-src."  At the tail end, the last 19 amino acids of the chicken protein "have been replaced with 12 unrelated residues."  The lost stretch held tyrosine 527, the site where the cell switches the protein off.  Changing that one tyrosine in the normal protein is enough to "induce cellular transformation."  Putting the undamaged chicken gene into the virus gave plenty of infectious virus and almost no tumours:  "overproduction of the c-src gene product does not cause cell transformation."^41421010

Two qualifications.  The viral gene also differs from the chicken gene at 7 to 17 other positions, and some of those changes activate it on their own.  And the virus without src is still a virus.  Borger makes that point himself.  The measured event is a virus picking up a host gene and breaking it, which fits Bergman's and Wood's claim well.  It is not a host gene turning into a virus.¹⁰¹⁶

Host jumps with dates

HIV.  A 1960 biopsy from Kinshasa, set beside a 1959 blood sample, dates the common ancestor of the pandemic strain "near the beginning of the 20th century."  HIV-1 crossed from apes to humans in "four independent cross-species transmission events."  See HIV Evolution for what the virus has done since.¹¹⁴³

Measles.  Measles virus is closest to rinderpest, a cattle virus.  Yuki Furuse dated their split to "around the 11th to 12th centuries."  Ariane Düx, adding a genome from a 1912 lung specimen, moved it to "528 BCE" with a range from 1174 BC to AD 165.  The two dates differ by some 1,600 years because they model the rate of change differently.  Both fall inside recorded history.  Nathan Wolfe, Claire Panosian Dunavan and Jared Diamond give the reason crowd diseases must be recent:  they "can only be sustained in large dense human populations."⁴⁴⁴⁵¹³

Myxoma.  This one was watched from the start.  The virus "causes a largely innocuous cutaneous fibroma in its natural host, the South American tapeti."  Released into Australia's European rabbits in 1950, it "had a case fatality rate (CFR) estimated at 99.8%."¹²¹²

Influenza and SARS-CoV-2 have their own articles.

Is a virus worse in a new host?

Often, and not always.  Ben Longdon injected one virus into 48 species of fruit fly.  The result was "benign infections in some species and rapid death in others," and related flies fared alike.  Sarah Guth found across animal viruses that reach humans that a more distant source species means more deaths and less spread.  Jemma Geoghegan found that host jumping "is a near universal feature" of the 19 virus families she compared.⁴⁶⁴⁷⁴⁸

The reservoir is not always safe either.  Monkey relatives of HIV were long thought harmless, and infected sooty mangabeys and African green monkeys "showed no sign of disease despite high viral loads."  Chimpanzees are different.  At Gombe, infected chimpanzees had "a 10- to 16-fold higher age-corrected death hazard."  The authors wrote that their findings "challenge the prevailing view that all natural SIV infections are non-pathogenic."  The chimpanzee virus is itself a blend of two monkey viruses, so the chimpanzee is a newer host too.  Rinderpest, the source of measles, was "a devastating cattle pathogen."  A host jump shows where a human virus came from.  It does not show that the virus was harmless there.^4949434313

Ebola's quiet relatives

Reston virus is an ebolavirus that kills macaques and has never been tied to human illness.  When it turned up in Philippine pigs in 2008, 6 of 141 people tested carried antibodies to it, and "there is no evidence of disease in humans despite the apparent occurrence of human infections."  Antibodies were found in 5 of 16 fruit bats of one Philippine species, and viral RNA in another.⁵⁰⁵¹⁵²

The African evidence is softer.  Thomas Monath argued in 1999 that "human contact with putative nonpathogenic EBO virus variants must occur frequently," because antibody surveys found far more exposure than disease.  His paper is subtitled "speculations," and he questioned the antibody tests himself.  A better survey in Gabon, published in 2010, found antibodies to Zaire ebolavirus, the lethal one, in 15.3 percent of 4,349 villagers.  A 2017 review of 51 surveys found little agreement on how often symptom-free infection occurs "or even if it exists."  A sixth ebolavirus, Bombali, was found in bats in 2018.  Its discoverers think it "possible that BOMV or some other potentially non-pathogenic ebolavirus has already spilled over."^53545556

So harmless ebolaviruses exist in the Philippines.  In Africa they are a reasonable inference that has not been confirmed by catching one in a person.

Prions

Prions are not viruses, but they follow the same pattern.  The protein behind scrapie and mad cow disease has a normal job.  Removing it from mice "leads to a chronic demyelinating polyneuropathy," a slow loss of the insulation around nerves.  Disease comes when the protein misfolds.  A different family of proteins uses prion-like chains at nerve junctions to hold long-term memories, which led Eric Kandel to ask, "Do you think God created prions just to kill?"  The two should not be confused.  The memory protein is not the disease protein.⁵⁷⁵⁸

Part 4: Testing the creationist proposals

Pathology is rare: passes, and proves little

Francis predicted that harmful microbes would be a small minority.  They are.  Most viruses infect bacteria, and most of the viruses in a person cause no illness.  But every view of virus origins expects this.  A parasite that kills all its hosts dies with them.  The prediction does not separate the models.

Harmless in the original host: often, not always

The tapeti and the sooty mangabey fit.  The Gombe chimpanzees and the cattle that died of rinderpest do not.  Thomas wrote in 2020 that HIV's relatives "play a normal role in the lives of several African monkey species."  He did not mention the chimpanzee study, published eleven years earlier.  He also wrote that "RNA virus genomes come integrated into their host genomes where they cause no harm."  That is true only of retroviruses.  Influenza, measles and coronaviruses do not insert into host DNA as part of their life cycle.⁴⁹²²²²

Sarfati suggests that bats' extra DNA-repair genes "may be due to a benign role of the virus."  No measurement supports a benefit.  Tolerating a virus and being helped by it are different things.²⁰

Escaped genes: mostly fails

Borger's claim is specific.  "It can be demonstrated that most RNA viruses are built of genetic information directly related to that of their hosts."  He offers the Rous sarcoma virus src gene, a part of HIV's coat protein, and influenza's neuraminidase.¹⁶

Four lines of evidence bear on it.

1. Hallmark genes. Koonin, Senkevich and Dolja found that the genes central to copying and packaging viruses "are shared by a broad variety of viruses but are missing from cellular genomes." In small RNA viruses the majority of genes "belong to the hallmark class." That is the opposite of Borger's statement.¹⁴¹⁴
2. Genes that match nothing. Yanbin Yin and Daniel Fischer searched 1,456 virus genomes. Of phage genes, "30.1% have no homologs neither in the viral nor in the prokaryotic world," and "almost half of the phage ORFs have prokaryotic homologs." A match does not show which way a gene travelled, since phages sit inside bacterial genomes. In ocean surveys "the majority (often 90%) of viral sequences are functionally 'unknown'".⁵⁹⁶⁰
3. Capsids. Here the model gets support. Mart Krupovic and Koonin compared the folded shapes of virus shell proteins and concluded that "many, if not all, capsid proteins evolved from ancestral proteins of cellular organisms on multiple, independent occasions." The resemblance is in shape, not in sequence, and they place the events in the deep past. It is still a cellular origin for the most virus-like part of a virus.⁶¹
4. The missing relatives. Liu and Soper name the problem in their own paper: "The biggest challenge to the idea of exogenization is that exogenous retroviruses include groups of complex retroviruses which have no endogenous counterparts." And: "there are no endogenous human lentiviruses, even though HIV is the best known exogenous human retrovirus."¹⁵

Where a virus has plainly taken a host gene, as with src and psbA, the gene is recognisable at once.  That is what makes the absence of a match for most viral genes informative.  The escaped-gene model may hold for some retroviruses.  As a general account of viruses it does not fit.  See Orphan Genes for the wider question of genes without relatives.

A recent origin: the clock agrees, with a caveat

One measured result favours the young-earth side.  Edward Holmes showed in 2003 that measured mutation rates imply "the families of RNA viruses circulating today could only have appeared very recently, probably not more than about 50,000 years ago."  Holmes called this a paradox and proposed that rates measured over decades cannot be stretched over long ages.  Both Borger and Thomas cite him.  The raw rates do point to recency.  Fifty thousand years is still eight times a biblical timescale, and the same clocks date measles and HIV comfortably inside it.⁶²

Objections and responses

"Viruses kill a fifth of ocean life daily.  That is carnage, not design."  The dead are bacteria and single-celled algae, and their contents feed the next generation of the same.  An ocean without that turnover would lock its nutrients in cells.  Whether regulation by killing is good design is a judgment.  That it regulates is a measurement.

"A designer who made phages to control cholera also made the phage that carries the toxin."  This is the strongest objection, and the young-earth writers who know the system best concede it is hard.  Their answer, that the toxin was made for a job inside sea animals, has no experiment behind it.  Until someone finds that job, the toxin is a well-made weapon with no known innocent use.

"Host jumps are evolution."  They are observed change, and no young-earth writer disputes them.  The measured cases are transfers, captures and losses.  src is a broken regulator.  Whether such changes can build new machinery is the subject of HIV Evolution, Influenza and SARS-CoV-2.

"Even the scientists disagree about when measles arose."  By 1,600 years.  Both dates are within the last 3,200 years.  Neither helps an argument that measles is an ancient companion of mankind.

"You quoted creationists against each other."  Liu and Soper and Francis and Wood state the weak points of their own models in print.  That is to their credit, and an honest summary has to repeat it.

A creationist objection:  "Beneficial viruses keep turning up, as we predicted."  They do, and the prediction was made before most of them were found.  But a benefit found today shows what a virus can do, not what it was first made for.  The parent article makes the same point about medical uses of parasites.

Issues and unanswered questions

• No harmless ancestor has been identified for most human viruses. Smallpox, polio, rabies and hepatitis B have no dated jump from a benign source in this article, because the sources read for it did not supply one. The young-earth account covers the cases with known origins and asserts the rest.
• The design answer has no explanation for viral weapons that fit human cells. Cholera toxin is the worked example. Francis and Wood say so.
• The escaped-gene model has never been tested by its authors. No creationist paper reports a systematic comparison of viral and host genes. The secular comparisons go mostly against it.
• The gut-regulator proposal conflicts with the one long-term gut study read here. Reyes found stable, quiet phage populations, not predator and prey.
• Several key papers were read only in abstract. Suttle's 2005 and 2007 reviews and the 1976 src paper are behind paywalls. The 20 percent figure is quoted from Suttle's 2013 lecture, not from the review usually cited for it.
• The African ebolavirus serology is unresolved. Cross-reaction between related viruses, mild infection with the lethal virus, and infection with an unknown harmless one all remain possible.
• Two studies with opposite implications come from the same oceans. Capsid shapes point to cells. Hallmark genes point away from them. Nobody on either side has shown how to weigh one against the other.

Verdict

Measured:  viruses are the most numerous biological objects on earth, most of them infect microbes, and they regulate microbial populations in the sea.  Some benefit their hosts, and one benefit has been proved by removing the virus and putting it back.  Every pathogenic virus in this article with a dated origin reached its present host by a jump from another species, or picked up and damaged a host gene.  In some of those cases the source host is unharmed.  In others it is not.  The gene capture and the harmless reservoirs are what Francis, Bergman and Wood expected.

Asserted:  that all viruses were once harmless, that Ebola helps bats, that gut phages guard the microbiome, and that viruses came out of host genomes.  The last is contradicted for most viral genes and conceded by its own authors to have no answer for HIV.

The design answer is weakest at the cholera toxin.  A corrupted regulator like src is easy to fit into a fallen world.  A toxin that enters human gut cells precisely, carried by a phage that delivers it, is not a broken thing.  The young-earth literature has named this problem and has not solved it.

Sources