How did chemicals first become organized into systems capable of self-propagation and adaptive evolution? One possibility is that the first evolvers were chemical ecosystems localized on mineral surfaces and composed of sets of molecular species that could catalyze each other’s formation. We used a bottom-up experimental framework, chemical ecosystem selection (CES), to evaluate this perspective and search for surface-associated and mutually catalytic chemical systems based on the changes in chemistry that they are expected to induce. Here, we report the results of preliminary CES experiments conducted using a synthetic “prebiotic soup” and pyrite grains, which yielded dynamical patterns that are suggestive of the emergence of mutual catalysis. While more research is needed to better understand the specific patterns observed here and determine whether they are reflective of self-propagation, these results illustrate the potential power of CES to test competing hypotheses for the emergence of protobiological chemical systems.

“Prebiotic soup” often features in discussions of origins of life research, both as a theoretical concept when discussing abiological pathways to modern biochemical building blocks and, more recently, as a feedstock in prebiotic chemistry experiments focused on discovering emergent, systems-level processes such as polymerization, encapsulation, and evolution. However, until now, little systematic analysis has gone into the design of well-justified prebiotic mixtures, which are needed to facilitate experimental replicability and comparison among researchers. This paper explores principles that should be considered in choosing chemical mixtures for prebiotic chemistry experiments by reviewing the natural environmental conditions that might have created such mixtures and then suggests reasonable guidelines for designing recipes. We discuss both “assembled” mixtures, which are made by mixing reagent grade chemicals, and “synthesized” mixtures, which are generated directly from diversity-generating primary prebiotic syntheses. We discuss different practical concerns including how to navigate the tremendous uncertainty in the chemistry of the early Earth and how to balance the desire for using prebiotically realistic mixtures with experimental tractability and replicability. Examples of two assembled mixtures, one based on materials likely delivered by carbonaceous meteorites and one based on spark discharge synthesis, are presented to illustrate these challenges. We explore alternative procedures for making synthesized mixtures using recursive chemical reaction systems whose outputs attempt to mimic atmospheric and geochemical synthesis. Other experimental conditions such as pH and ionic strength are also considered. We argue that developing a handful of standardized prebiotic recipes may facilitate coordination among researchers and enable the identification of the most promising mechanisms by which complex prebiotic mixtures were “tamed” during the origin of life to give rise to key living processes such as self-propagation, information processing, and adaptive evolution. We end by advocating for the development of a public prebiotic chemistry database containing experimental methods (including soup recipes), results, and analytical pipelines for analyzing complex prebiotic mixtures.

Many attempts to ascertain the physicochemical processes governing the emergence of life have focused on studying the synthesis of particular biomolecules and their precursors or on designing simple systems that manifest life-like properties. In this paper, we present the methodological components of an experimental framework designed to generate and detect spontaneously forming chemical systems capable of collective propagation and adaptive evolution. The chemical ecosystem selection paradigm involves incubating complex mixtures of organic compounds with populations of mineral grains to promote the appearance of autocatalytic sets that interact with the mineral surface and perform serial transfers to favor surface-associated systems that are better at being transmitted from grain to grain. This approach has the potential to serve as a novel tool for screening a vast array of experimental conditions and determine the likelihood that they will produce life-like chemical systems. Many attempts to ascertain the physicochemical processes governing the emergence of life have focused on studying the synthesis of particular biomolecules and their precursors or on designing simple systems that manifest life-like properties. In this paper, we present the methodological components of an experimental framework designed to generate and detect spontaneously forming chemical systems capable of collective propagation and adaptive evolution. The chemical ecosystem selection paradigm involves incubating complex mixtures of organic compounds with populations of mineral grains to promote the appearance of autocatalytic sets that interact with the mineral surface and perform serial transfers to favor surface-associated systems that are better at being transmitted from grain to grain. This approach has the potential to serve as a novel tool for screening a vast array of experimental conditions and determine the likelihood that they will produce life-like chemical systems. Many attempts to ascertain the physicochemical processes governing the emergence of life have focused on studying the synthesis of particular biomolecules and their precursors or on designing simple systems that manifest life-like properties. In this paper, we present the methodological components of an experimental framework designed to generate and detect spontaneously forming chemical systems capable of collective propagation and adaptive evolution. The chemical ecosystem selection paradigm involves incubating complex mixtures of organic compounds with populations of mineral grains to promote the appearance of autocatalytic sets that interact with the mineral surface and perform serial transfers to favor surface-associated systems that are better at being transmitted from grain to grain. This approach has the potential to serve as a novel tool for screening a vast array of experimental conditions and determine the likelihood that they will produce life-like chemical systems.