Two recent experiments highlight the power of mind to direct natural processes against astronomical improbabilities. One study boasts of getting “New genes out of nothing.” (The echo of Lawrence Krauss’s book A Universe from Nothing is noteworthy.) The other study shows what human minds can do with protein precursors (the translated products of genes) by using “design selection” as opposed to natural selection.
In the “genes out of nothing” camp, researchers at Uppsala University describe the challenge they face — and that standard Darwinian processes had to face:
How do new genes and functional proteins arise and develop? This is one of the most fundamental issues in evolutionary biology. Two different types of mechanism have been proposed: (1) new genes with novel functions arise from existing genes, and (2) new genes and proteins evolve from random DNA sequences with no similarity to existing genes and proteins. In the present study, the researchers explored the latter type of mechanism: evolution of new genes and proteins from randomised DNA sequences – de novo evolution, as it is called. It is fairly easy to understand that when a gene already exists, it can be modified and acquire a new function. But how does “nothing” turn into a function affording a small advantage that is favoured by natural selection? [Emphasis added.]
Upon reading this news, however, one finds another case of artificial selection. True, they started with 500 million randomized sequences, but the scientists selected the goal: finding polypeptides able to confer antibiotic resistance. While they were delighted to see successes out of their sample, their experiment had nothing to do with natural selection. Even worse, they only tested short polypeptides 22-25 amino acid residues long. The sequence space rises exponentially with polypeptide length, ensuring that even small proteins of 100 aa would be vastly outnumbered by useless ones.
Also, if they had started with racemic amino acids (a mixture of left- and right-handed ones), the improbability of finding workable solutions would have dropped through the floor. Their open-access paper in the journal mBio is overly optimistic with unverified hope:
Here, we present experimental selection [i.e., artificial selection] for peptides that confer a beneficial functionality in vivo and that were generated from randomized nucleotide sequences, supporting the idea that expression of randomly occurring sORFs [small open reading frames] can serve as a substrate for evolution of de novo genes. Further work is needed to determine what fraction of sequence space can generate such beneficial functions and which factors constrain how rapidly a proto-peptide/protein can be fine-tuned in its cellular function and how the initial fitness costs can be reduced. It would be interesting to examine if a cell can evolve to reduce the deleterious side effects of Arp expression. Evolution of such variants would likely require multiple genetic alterations which could be achieved with large bacterial populations during long-term evolution experiments in laboratory settings.
Those questions have already been answered. If they had checked Douglas Axe’s book, Undeniable (Chapter 8, and pp. 180-181), they would have found empirical proof that the fraction of sequence space that is functional for a small protein is nanoscopically minute. If they had checked Michael Behe’s book, Darwin Devolves (Chapters 7-8), they would have learned how Richard Lenski’s Long Term Evolution Experiment and Joseph Thornton’s receptor experiment have already demonstrated the paltry benefits of natural selection under ideal lab settings. In every case, presumed benefits occurred by breaking or blunting existing genetic information.
Further reading of the paper shows numerous examples of investigator interference, to say nothing of question-begging the prior existence of established systems for transcription and translation. The details, therefore, undermine the flashy headline, “New genes from nothing.”
Minds on the Matter
In a very different study, citizen scientists were challenged to run directed searches through random sequences to find folding proteins. The goal of this project, published in Nature (“De novo protein design by citizen scientists”), was to design polypeptides able to fold into pre-determined shapes.
We posed the challenge of de novo protein design in the online protein-folding game Foldit. Players were presented with a fully extended peptide chain and challenged to craft a folded protein structure and an amino acid sequence encoding that structure. After many iterations of player design, analysis of the top-scoring solutions and subsequent game improvement, Foldit players can now — starting from an extended polypeptide chain — generate a diversity of protein structures and sequences that encode them in silico.
Included in some of the 146 winning solutions were some 20 folds, one of them not found in nature. Out of the solution sequences, 56 were tested in E. coli and found to fold just as the players had designed them. The experiment shows the superiority of minds in finding solutions to problems of this sort, even when the participants lack expertise in protein modeling.
This work makes explicit the considerable implicit knowledge that contributes to success in de novo protein design, and shows that citizen scientists can discover creative new solutions to outstanding scientific challenges such as the protein design problem.
The paper never mentions natural selection, evolution, or mutations. The word design, however, appears well over 150 times in the main paper, with more in the Supplementary Information document. The authors are aware of the futility of searching sequence space without a plan:
The principle underlying de novo protein design is that proteins fold to their lowest free-energy state; hence, designing a new protein structure requires finding an amino acid sequence with its lowest energy state in the prescribed structure. In practice, this challenge can be divided into two subproblems: first, crafting a protein backbone that is designable (that is, that could be the lowest energy state of some sequence); and second, finding a sequence with its lowest energy state in the crafted structure. One of the challenges of protein design is the exponentially increasing number of conformations available to a polypeptide chain, which is huge even for a modestly sized protein of 60–100 residues.
Expecting a “blind watchmaker” to wade through the huge space of possible sequences is reminiscent of William Dembski’s analogy of a blindfolded treasure hunter digging at random on an island without a treasure map (No Free Lunch, Chapter 4). Watchmakers (or protein-makers in this case) with eyes open are able to cut to the chase rapidly by having a goal in mind and a strategy to reach it. A creative mind can filter out blind alleys and learn from mistakes. And citizen scientists have a gift called creativity — a feature of intelligence — that compensates for lack of expert knowledge.
Disentangling the role of expert knowledge is particularly difficult for the extremely open-ended challenge posed by the first subproblem (that is, crafting a plausible backbone), for which there are a practically unlimited number of solutions. Because full computer enumeration of backbones is not possible, there is considerable room for human creativity and intuition in generating and designing new protein structures.
Because computer games are popular, and many online gamers love a challenge, this experiment unleashed a torrent of creativity. The study authors even had ways to watch how the solutions were being achieved over time, because Foldit uploaded their working models every few minutes. Participants tried to outdo each other in playing Foldit to find solutions within the seven-day time limit. The most successful player created ten successful designs. Natural selection (the blind protein-maker) would have required more time than the age of the universe to do what intelligent agents accomplished within a week.
Two recent studies demonstrate the dramatic superiority of intelligence in solving the problem of getting form or function out of randomness. Out of a vast number of possible sequences, which polypeptides can fold and form new structures? The first study claimed that evolution could do it from nothing, but the investigators interfered by using existing machinery and manipulating the random sequences toward their own predetermined goal. The second study turned loose the creative intelligence of citizen scientists, who rapidly found numerous solutions against overwhelming odds. When it comes to natural selection versus design selection, there’s no contest.
Photo: DNA researchers at work, by Fred W. Baker III via U.S. Department of Defense.