Darwinian Selection Process Research Articles

450 million years in the making: mapping the evolutionary foundations of germinal centers.

Germinal centers (GCs) are distinct microanatomical structures that form in the secondary lymphoid organs of endothermic vertebrates (i.e., mammals and some birds). Within GCs, B cells undergo a Darwinian selection process to identify clones which can respond to pathogen insult as well as affinity mature the B cell repertoire. The GC response ultimately generates memory B cells and bone marrow plasma cells which facilitate humoral immunological memory, the basis for successful vaccination programs. GCs have not been observed in the secondary lymphoid organs of ectothermic jawed vertebrates (i.e., fishes, reptiles, and amphibians). However, abundant research over the past decades has indicated these organisms can produce antigen specific B cell responses and some degree of affinity maturation. This review examines data demonstrating that the fundamentals of B cell selection may be more conserved across vertebrate phylogeny than previously anticipated. Further, research in both conventional mammalian model systems and comparative models raises the question of what evolutionary benefit GCs provide endotherms if they are seemingly unnecessary for generating the basic functional components of jawed vertebrate humoral adaptive immune responses.

Institutions, assortative matching and cultural evolution

This paper studies the role of institution in a Darwinian evolutionary process of cultural selection. The primary function of an institution is to determine how citizens in a society are matched pairwisely to interact. We examine three different types of institutions: utilitarian, egalitarian, and Nash. Two cultural types stand out in the long run through the evolutionary process. The opportunistic cultural type maximizes individual payoff against another opportunistic cultural type, while the civic-minded cultural type maximizes the total payoff of a pair. We show that the structure of the underlying interactions among citizens plays a critical role. On the one hand, the evolutionary stability of the civic-minded cultural type requires supermodularity of the citizens’ payoff function under the utilitarian institution and log-supermodularity in addition under the Nash institution. On the other hand, the evolutionary stability of the opportunistic cultural type requires submodularity of the citizens’ payoff function under the utilitarian institution and log-submodularity under the Nash institution. Neither type’s evolutionary stability is guaranteed under the egalitarian institution.

Quantifying the evolvability of allostery at different protein surfaces.

Allosteric regulation requires the cooperative action of multiple amino acid residues to relay a signal from a distal surface site to the protein active site. It is an open problem to understand how new regulation evolves through a Darwinian process of single mutation and selection. Yet a better understanding of how allostery emerges is critical to our ability to engineer new custom biosensors. Are some protein surfaces more prone to evolve into allosteric sites than others? Or is the capacity to evolve allostery homogenously distributed across a protein's entire surface? Recent work has identified that allosteric regulation is more easily engineered at certain protein surfaces termed “allosteric hotspots.” These hotspots occur at surface positions that co-evolve with enzyme active sites. Allosteric hotspots have been used to engineer new allosteric regulation of dihydrofolate reductase (DHFR) by the light-oxygen-voltage-sensing domain (LOV2). When LOV2 is fused to allosteric hotspots on DHFR, DHFR activity is weakly regulated by light (1.3-fold). However, fusing LOV2 to non-hotspot positions does not result in allosteric control. The low level of regulation observed in DHFR/LOV2 fusions presents an opportunity to study how allostery can be optimized by evolution. In my work, I am using random mutagenesis combined with a high-throughput screen to evolve new allosteric regulation. I am performing forward evolution of allostery for a weakly allosteric, hotspot-connected fusion, and for a non-allosteric, non-hotspot-connected fusion. I am monitoring the distribution of mutational effects on allostery for each construct over rounds of selection. With this approach, I will quantify the evolvability of allosteric regulation at different protein surface sites. I hypothesize that allosteric hotspots will have a higher prevalence of allostery-tuning mutations, and that these mutations will result in higher levels of allosteric regulation.

Germinal Centers.

Germinal centers (GCs) are microanatomical sites of B cell clonal expansion and antibody affinity maturation. Therein, B cells undergo the Darwinian process of somatic diversification and affinity-driven selection of immunoglobulins that produces the high-affinity antibodies essential for effective humoral immunity. Here, we review recent developments in the field of GC biology, primarily as it pertains to GCs induced by infection or immunization. First, we summarize the phenotype and function of the different cell types that compose the GC, focusing on GC B cells. Then, we review the cellular and molecular bases of affinity-dependent selection within the GC and the export of memory and plasma cells. Finally, we present an overview of the emerging field of GC clonal dynamics, focusing on how GC and post-GC selection shapes the diversity of antibodies secreted into serum.

Tumor Heterogeneity and Consequences for Bladder Cancer Treatment.

Simple SummaryBladder cancer is a heterogeneous disease that is composed of epithelia with varying transcriptional, mutational and lineage signatures. The epithelia of bladder tumors can also undergo pronounced changes in transcriptional and phenotypical qualities in response to progression, treatment related stresses and cues from the tumor microenvironment (TME). We hypothesize that changes in epithelial tumor heterogeneity (EpTH) occur due to the evolving content of epithelial subpopulations through both Darwinian and Lamarckian-like natural selection processes. We further conjecture that lineage-defined subpopulations can change through nongenomic and genomic cellular mechanisms that include cellular plasticity and acquired driver mutations, respectively. We propose that such processes are dynamic and contribute towards clinical treatment challenges including progression to drug resistance. In this article, we assess mechanisms that may support dynamic tumor heterogeneity with the overall goal of emphasizing the application of these concepts to the clinical setting.Acquired therapeutic resistance remains a major challenge in cancer management and associates with poor oncological outcomes in most solid tumor types. A major contributor is tumor heterogeneity (TH) which can be influenced by the stromal; immune and epithelial tumor compartments. We hypothesize that heterogeneity in tumor epithelial subpopulations—whether de novo or newly acquired—closely regulate the clinical course of bladder cancer. Changes in these subpopulations impact the tumor microenvironment including the extent of immune cell infiltration and response to immunotherapeutics. Mechanisms driving epithelial tumor heterogeneity (EpTH) can be broadly categorized as mutational and non-mutational. Mechanisms regulating lineage plasticity; acquired cellular mutations and changes in lineage-defined subpopulations regulate stress responses to clinical therapies. If tumor heterogeneity is a dynamic process; an increased understanding of how EpTH is regulated is critical in order for clinical therapies to be more sustained and durable. In this review and analysis, we assess the importance and regulatory mechanisms governing EpTH in bladder cancer and the impact on treatment response.

Biogeomorphological eco-evolutionary feedback between life and geomorphology: a theoretical framework using fossorial mammals.

Engineer organisms not only adapt to pre-existing environmental conditions but also co-construct their physical environment. By doing so, they can subsequently change selection pressures for themselves and other species, as well as change community and ecosystem structures and functions. Focusing on one representative example, i.e., fossorial mammals, we show that geomorphological Earth system components are crucial for understanding and quantifying links between evolutionary and ecosystem dynamics and that feedbacks between geomorphology and engineer organisms constitute a major driver of geomorphological organization on the Earth's surface. We propose a biogeomorphological eco-evolutionary feedback synthesis from the gene to the landscape where eco-evolutionary feedbacks are mediated by the geomorphological dimensions of a niche that are affected by engineer organisms, such as fossorial mammals. Our concept encompasses (i) the initial responses of fossorial mammals to environmental constraints that enhance the evolution of their morphological and biomechanical traits for digging in the soil; (ii) specific adaptations of engineer fossorial mammals (morphological, biomechanical, physiological and behavioural feedback traits for living in burrows) to their constructed geomorphological environment; and (iii) ecological and evolutionary feedbacks diffusing at the community and ecological levels. Such a new perspective in geomorphology may lead to a better conceptualization and analysis of Earth surface processes and landforms as parts of complex adaptive systems in which Darwinian selection processes at lower landscape levels lead to self-organization of higher-level landforms and landscapes.

Recurrent chromosomal imbalances provide selective advantage to human embryonic stem cells under enhanced replicative stress conditions.

Human embryonic stem cells (hESCs) and embryonal tumors share a number of common features, including a compromised G1/S checkpoint. Consequently, these rapidly dividing hESCs and cancer cells undergo elevated levels of replicative stress, inducing genomic instability that drives chromosomal imbalances. In this context, it is of interest that long-term in vitro cultured hESCs exhibit a remarkable high incidence of segmental DNA copy number gains, some of which are also highly recurrent in certain malignancies such as 17q gain (17q+). The selective advantage of DNA copy number changes in these cells has been attributed to several underlying processes including enhanced proliferation. We hypothesized that these recurrent chromosomal imbalances become rapidly embedded in the cultured hESCs through a replicative stress driven Darwinian selection process. To this end, we compared the effect of hydroxyurea-induced replicative stress vs normal growth conditions in an equally mixed cell population of isogenic euploid and 17q + hESCs. We could show that 17q + hESCs rapidly overtook normal hESCs. Our data suggest that recurrent chromosomal segmental gains provide a proliferative advantage to hESCs under increased replicative stress, a process that may also explain the highly recurrent nature of certain imbalances in cancer.

Abstract PO-018: Single cell whole genome sequencing reveals the dynamics of copy number instability at the earliest stages of cancer evolution

Abstract Background: Cancer evolution occurs though a Darwinian process of natural selection. Selective sweeps occur in most human tumours which obscure early, unsuccessful lineages. In addition, analysis of bulk samples has limited phylogenies to SNVs and Indels, due to methodological difficulties in bulk clonal deconvolution of copy number aberrations (CNAs). Using single cell whole genome sequencing we construct phylogenies of CNAs for murine models of the early stages in lung adenocarcinoma evolution and define features of unsuccessful lineages not yet obscured by full selective sweeps. Methods: Non-invasive tumour tissue was homogenised from EGFRL858R and EGFRL858R/TP53 knock out (KO) mice and subject to 0.1X single cell WGS. Bowtie and AnneuFinder were used to align FASTQ files and call CNAs respectively. A strict quality control step was step applied using read depth and coverage variability, removing 15% of cells. MEDDIC was used to construct phylogenetic trees and ancestral CN states at each node. These ancestral CN states were translated into distinct CN events in the evolutionary history of each tumour. Results: We reconstructed tumour CNA phylogenies using scWGS to determine phylogenetic relationships and time CN events. CNAs in these data were far more heterogenous than in previously published human studies. We noted small populations of progenitor cells which branched off early in evolutionary time, containing far fewer copy number aberrations and which were not genome doubled. These populations allow us to more accurately time CN events such as would be obscured after a full selective sweep. Whole chromosomal events were significantly more likely to occur earlier in evolutionary history than intra-chromosomal events (Wilcoxon test, P < 1e-16, OR = 2.24) and chromosomal gains were also significantly more common at the earlier stages of tumour evolution (Wilcoxon test, P < 1e-16, OR = 1.44). Within whole chromosomal events, gains were significantly more enriched for early timing (Wilcoxon test, P < 1e-16, OR = 7.11). Progenitor cell populations also contained a small number of additional events, which were usually unique to a single cell. We reasoned these events may represent evolutionary dead ends. We found these events were highly enriched for losses in TP53 wild type tumours (Wilcoxon test, P < 1e-16, OR = 11.9) but much less so in TP53 mutant tumours (Wilcoxon test, P = 1e-6, OR = 1.14) suggesting early negative selection against loss of heterozygosity, particularly when TP53 is intact. Discussion: The observed depletion of intrachromosomal events may be caused by a more intact DNA damage response system which is therefore less permissive to chromosomal breaks. Heterozygous losses may be negatively selected due to resultant decreased dosage of a large numbers of genes, upon which the cells become less reliant as cancer progresses. These data are consistent with a recent report of spontaneous CNAs in culture of normal hTERT lines which have a strong preference for gains in a TP53 WT context but not upon TP53 KO. Citation Format: Alexander M. Frankell, Sebastian Hobor, Maise Al-bakir, Eva Grönroos, Charles Swanton. Single cell whole genome sequencing reveals the dynamics of copy number instability at the earliest stages of cancer evolution [abstract]. In: Proceedings of the AACR Virtual Special Conference on Tumor Heterogeneity: From Single Cells to Clinical Impact; 2020 Sep 17-18. Philadelphia (PA): AACR; Cancer Res 2020;80(21 Suppl):Abstract nr PO-018.

A generalized theory of age-dependent carcinogenesis.

This article has been through an editorial process in which the authors decide how to respond to the issues raised during peer review. The Reviewing Editor's assessment is that all the issues have been addressed (see decision letter).

Benchmarking Tree and Ancestral Sequence Inference for B Cell Receptor Sequences.

B cell receptor sequences evolve during affinity maturation according to a Darwinian process of mutation and selection. Phylogenetic tools are used extensively to reconstruct ancestral sequences and phylogenetic trees from affinity-matured sequences. In addition to using general-purpose phylogenetic methods, researchers have developed new tools to accommodate the special features of B cell sequence evolution. However, the performance of classical phylogenetic techniques in the presence of B cell-specific features is not well understood, nor how much the newer generation of B cell specific tools represent an improvement over classical methods. In this paper we benchmark the performance of classical phylogenetic and new B cell-specific tools when applied to B cell receptor sequences simulated from a forward-time model of B cell receptor affinity maturation toward a mature receptor. We show that the currently used tools vary substantially in terms of tree structure and ancestral sequence inference accuracy. Furthermore, we show that there are still large performance gains to be achieved by modeling the special mutation process of B cell receptors. These conclusions are further strengthened with real data using the rules of isotype switching to count possible violations within each inferred phylogeny.

Does Evolution Conflict with God's Character?

AbstractThe core question that this article addresses is how the Darwinian evolutionary process of random mutation and natural selection relates to God's character, especially as portrayed in the Christian tradition. The main challenge is that the process of random mutation and natural selection generally favours well‐adapted individuals, whereas, in the Christian tradition, God is said to favour the weak and poor. I first spell out the problem in more detail and explain how it is different from other problems regarding the evolutionary process and God's character that can be found in the philosophical and theological literature. Subsequently, I draw attention to various misunderstandings, especially about the nature of Darwinian evolution. Doing so weakens the tension between Darwinian evolution and God's character, but does not provide a satisfactory solution. I therefore move on to discuss six attempts to solve the problem and argue that each of them is seriously wanting. Finally, I propose three lines of reasoning that jointly constitute a sufficiently plausible solution to the problem.

Comparison of Burnet's clonal selection theory with tumor cell-clone development.

Increasing evidence has shown that Darwin's theory of evolution provides vital insights into the emergence and etiology of different types of cancer. On a microscopic scale, cancer stem cells meet the conditions for the Darwinian process of natural selection. In particular, cancer stem cells undergo cell reproduction characterized by the emergence of heritable variability that promotes replication and cell survival.Methods: Evidence from previous studies was gathered to compare Burnet's clonal selection theory with the tumor evolution theory.Results: The findings show that the Darwinian theory offers a general framework for understanding fundamental aspects of cancer. As fundamental theoretical frameworks, Burnet's clonal selection theory and the tumor evolution theory can be used to explain cancer cell evolution and identify the beneficial adaptations that contribute to cell survival in tissue landscapes and tissue ecosystems.Conclusions: In conclusion, this study shows that both Burnet's clonal selection theory and the tumor evolution theory postulate that cancer cells in tissue ecosystems evolve through reiterative processes, such as clonal expansion, clonal selection, and genetic diversification. Therefore, both theories provide insights into the complexities and dynamics of cancer, including its development and progression. Finally, we take into account the occurrence of biologic variation in both tumor cells and lymphocytes. It is important to note that the presence of lymphocyte variations appears to be advantageous in the framework of tumor defense but also dangerous within the framework of autoimmune disease development.

Intraspecific eye color variability in birds and mammals: a recent evolutionary event exclusive to humans and domestic animals

BackgroundHuman populations and breeds of domestic animals are composed of individuals with a multiplicity of eye (= iris) colorations. Some wild birds and mammals may have intraspecific eye color variability, but this variation seems to be due to the developmental stage of the individual, its breeding status, and/or sexual dimorphism. In other words, eye colour tends to be a species-specific trait in wild animals, and the exceptions are species in which individuals of the same age group or gender all develop the same eye colour. Domestic animals, by definition, include bird and mammal species artificially selected by humans in the last few thousand years. Humans themselves may have acquired a diverse palette of eye colors, likewise in recent evolutionary time, in the Mesolithic or in the Upper Paleolithic.Presentation of the hypothesisWe posit two previously unrecognized hypotheses regarding eye color variation: 1) eye coloration in wild animals of every species tends to be a fixed trait. 2) Humans and domestic animal populations, on the contrary, have eyes of multiple colors. Sexual selection has been invoked for eye color variation in humans, but this selection mode does not easily apply in domestic animals, where matings are controlled by the human breeder.Testing the hypothesisEye coloration is polygenic in humans. We wish to investigate the genetics of eye color in other animals, as well as the ecological correlates.Implications of the hypothesisInvestigating the origin and function of eye colors will shed light on the reason why some species may have either light-colored irises (e.g., white, yellow or light blue) or dark ones (dark red, brown or black). The causes behind the vast array of eye colors across taxa have never been thoroughly investigated, but it may well be that all Darwinian selection processes are at work: sexual selection in humans, artificial selection for domestic animals, and natural selection (mainly) for wild animals.

Mendel in the C-Suite: Design and the Evolution of Strategies

A “Mendelian” executive is proposed as an image of strategy making that lies intermediate between the godlike powers of intentional design of rational choice approaches and a Darwinian process of random variation and market-based differential selection. The Mendelian executive is capable of intentional design efforts in order to explore possible adjacent strategic spaces. Furthermore, the argument developed here highlights the role of intentionality with respect to the selection and culling of strategic initiatives. The firm is viewed as operating an “artificial selection” environment in contrast to selection as the direct consequence of the outcome of competitive processes. Examining the nature of the processes generating these experimental variants and the bases of internal selection, and how these selection criteria may themselves change, is argued to be central to the formation of strategy in dynamic competitive environments.

Evolutionary biology of high-risk multiple myeloma.

The outcomes for the majority of patients with myeloma have improved over recent decades, driven by treatment advances. However, there is a subset of patients considered to have high-risk disease who have not benefited. Understanding how high-risk disease evolves from more therapeutically tractable stages is crucial if we are to improve outcomes. This can be accomplished by identifying the genetic mechanisms and mutations driving the transition of a normal plasma cell to one with the features of the following disease stages: monoclonal gammopathy of undetermined significance, smouldering myeloma, myeloma and plasma cell leukaemia. Although myeloma initiating events are clonal, subsequent driver lesions often occur in a subclone of cells, facilitating progression by Darwinian selection processes. Understanding the co-evolution of the clones within their microenvironment will be crucial for therapeutically manipulating the process. The end stage of progression is the generation of a state associated with treatment resistance, increased proliferation, evasion of apoptosis and an ability to grow independently of the bone marrow microenvironment. In this Review, we discuss these end-stage high-risk disease states and how new information is improving our understanding of their evolutionary trajectories, how they may be diagnosed and the biological behaviour that must be addressed if they are to be treated effectively.

An evolutionary perspective on the systems of adaptive immunity.

We propose an evolutionary perspective to classify and characterize the diverse systems of adaptive immunity that have been discovered across all major domains of life. We put forward a new function-based classification according to the way information is acquired by the immune systems: Darwinian immunity (currently known from, but not necessarily limited to, vertebrates) relies on the Darwinian process of clonal selection to 'learn' by cumulative trial-and-error feedback; Lamarckian immunity uses templated targeting (guided adaptation) to internalize heritable information on potential threats; finally, shotgun immunity operates through somatic mechanisms of variable targeting without feedback. We argue that the origin of Darwinian (but not Lamarckian or shotgun) immunity represents a radical innovation in the evolution of individuality and complexity, and propose to add it to the list of major evolutionary transitions. While transitions to higher-level units entail the suppression of selection at lower levels, Darwinian immunity re-opens cell-level selection within the multicellular organism, under the control of mechanisms that direct, rather than suppress, cell-level evolution for the benefit of the individual. From a conceptual point of view, the origin of Darwinian immunity can be regarded as the most radical transition in the history of life, in which evolution by natural selection has literally re-invented itself. Furthermore, the combination of clonal selection and somatic receptor diversity enabled a transition from limited to practically unlimited capacity to store information about the antigenic environment. The origin of Darwinian immunity therefore comprises both a transition in individuality and the emergence of a new information system-the two hallmarks of major evolutionary transitions. Finally, we present an evolutionary scenario for the origin of Darwinian immunity in vertebrates. We propose a revival of the concept of the 'Big Bang' of vertebrate immunity, arguing that its origin involved a 'difficult' (i.e. low-probability) evolutionary transition that might have occurred only once, in a common ancestor of all vertebrates. In contrast to the original concept, we argue that the limiting innovation was not the generation of somatic diversity, but the regulatory circuitry needed for the safe operation of amplifiable immune responses with somatically acquired targeting. Regulatory complexity increased abruptly by genomic duplications at the root of the vertebrate lineage, creating a rare opportunity to establish such circuitry. We discuss the selection forces that might have acted at the origin of the transition, and in the subsequent stepwise evolution leading to the modern immune systems of extant vertebrates.

Clonal selection and asymmetric distribution of human leukemia in murine xenografts revealed by cellular barcoding

AbstractGenetic and phenotypic heterogeneity of human leukemia is thought to drive leukemia progression through a Darwinian process of selection and evolution of increasingly malignant clones. However, the lack of markers that uniquely identify individual leukemia clones precludes high-resolution tracing of their clonal dynamics. Here, we use cellular barcoding to analyze the clonal behavior of patient-derived leukemia-propagating cells (LPCs) in murine xenografts. Using a leukemic cell line and diagnostic bone marrow cells from 6 patients with B-progenitor cell acute lymphoblastic leukemia, we demonstrate that patient-derived xenografts were highly polyclonal, consisting of tens to hundreds of LPC clones. The number of clones was stable within xenografts but strongly reduced upon serial transplantation. In contrast to primary recipients, in which clonal composition was highly diverse, clonal composition in serial xenografts was highly similar between recipients of the same donor and reflected donor clonality, supporting a deterministic, clone-size–based model for clonal selection. Quantitative analysis of clonal abundance in several anatomic sites identified 2 types of anatomic asymmetry. First, clones were asymmetrically distributed between different bones. Second, clonal composition in the skeleton significantly differed from extramedullary sites, showing similar numbers but different clone sizes. Altogether, this study shows that cellular barcoding and xenotransplantation providea useful model to study the behavior of patient-derived LPC clones, which provides insights relevant for experimental studies on cancer stem cells and for clinical protocols for the diagnosis and treatment of leukemia.

Likelihood-Based Inference of B Cell Clonal Families.

The human immune system depends on a highly diverse collection of antibody-making B cells. B cell receptor sequence diversity is generated by a random recombination process called “rearrangement” forming progenitor B cells, then a Darwinian process of lineage diversification and selection called “affinity maturation.” The resulting receptors can be sequenced in high throughput for research and diagnostics. Such a collection of sequences contains a mixture of various lineages, each of which may be quite numerous, or may consist of only a single member. As a step to understanding the process and result of this diversification, one may wish to reconstruct lineage membership, i.e. to cluster sampled sequences according to which came from the same rearrangement events. We call this clustering problem “clonal family inference.” In this paper we describe and validate a likelihood-based framework for clonal family inference based on a multi-hidden Markov Model (multi-HMM) framework for B cell receptor sequences. We describe an agglomerative algorithm to find a maximum likelihood clustering, two approximate algorithms with various trade-offs of speed versus accuracy, and a third, fast algorithm for finding specific lineages. We show that under simulation these algorithms greatly improve upon existing clonal family inference methods, and that they also give significantly different clusters than previous methods when applied to two real data sets.

Just like the rest of evolution in Mother Nature, the evolution of cancers may be driven by natural selection, and not by haphazard mutations.

Sporadic carcinogenesis starts from immortalization of a differentiated somatic cell or an organ-specific stem cell. The immortalized cell incepts a new or quasinew organism that lives like a parasite in the patient and usually proceeds to progressive simplification, constantly engendering intermediate organisms that are simpler than normal cells. Like organismal evolution in Mother Nature, this cellular simplification is a process of Darwinian selection of those mutations with growth- or survival-advantages, from numerous ones that occur randomly and stochastically. Therefore, functional gain of growth- or survival-sustaining oncogenes and functional loss of differentiation-sustaining tumor suppressor genes, which are hallmarks of cancer cells and contribute to phenotypes of greater malignancy, are not drivers of carcinogenesis but are results from natural selection of advantageous mutations. Besides this mutation-load dependent survival mechanism that is evolutionarily low and of an asexual nature, cancer cells may also use cell fusion for survival, which is an evolutionarily-higher mechanism and is of a sexual nature. Assigning oncogenes or tumor suppressor genes or their mutants as drivers to induce cancer in animals may somewhat coerce them to create man-made oncogenic pathways that may not really be a course of sporadic cancer formations in the human.

THERMAL SOLUTIONS FOR MOLECULAR EVOLUTION

The key requirement to solve the origin of life puzzle are disequilibrium conditions. Early molecular evolution cannot be explained by initial high concentrations of energetic chemicals since they would just react towards their chemical equilibrium allowing no further development. We argue here that persistent disequilibria are needed to increase complexity during molecular evolution. We propose thermal gradients as the disequilibrium setting which drove Darwinian molecular evolution. On the one hand the thermal gradient gives rise to laminar thermal convection flow with highly regular temperature oscillations that allow melting and replication of DNA. On the other hand molecules move along the thermal gradient, a mechanism termed Soret effect or thermophoresis. Inside a long chamber a combination of the convection flow and thermophoresis leads to a very efficient accumulation of molecules. Short DNA is concentrated thousand-fold, whereas longer DNA is exponentially better accumulated. We demonstrated both scenarios in the same micrometer-sized setting. Forthcoming experiments will reveal how replication and accumulation of DNA in a system, driven only by a thermal gradient, could create a Darwinian process of replication and selection.