Preserved soft anatomy confirms shoulder-powered upstroke of early theropod flyers, reveals enhanced early pygostylian upstroke, and explains early sternum loss.

Anatomy of the first flying feathered dinosaurs, modern birds and crocodylians, proposes an ancestral flight system divided between shoulder and chest muscles, before the upstroke muscles migrated beneath the body. This ancestral flight system featured the dorsally positioned deltoids and supracoracoideus controlling the upstroke and the chest-bound pectoralis controlling the downstroke. Preserved soft anatomy is needed to contextualize the origin of the modern flight system, but this has remained elusive. Here we reveal the soft anatomy of the earliest theropod flyers preserved as residual skin chemistry covering the body and delimiting its margins. These data provide preserved soft anatomy that independently validate the ancestral theropod flight system. The heavily constructed shoulder and more weakly constructed chest in the early pygostylian Confuciusornis indicated by a preserved body profile, proposes the first upstroke-enhanced flight stroke. Slender ventral body profiles in the early-diverging birds Archaeopteryx and Anchiornis suggest habitual use of the pectoralis could not maintain the sternum through bone functional adaptations. Increased wing-assisted terrestrial locomotion potentially accelerated sternum loss through higher breathing requirements. Lower expected downstroke requirements in the early thermal soarer Sapeornis could have driven sternum loss through bone functional adaption, possibly encouraged by the higher breathing demands of a Confuciusornis-like upstroke. Both factors are supported by a slender ventral body profile. These data validate the ancestral shoulder/chest flight system and provide insights into novel upstroke-enhanced flight strokes and early sternum loss, filling important gaps in our understanding of the appearance of modern flight.

Response to Serrano and Chiappe.

In the recent study in Current Biology by Pei and colleagues1, we used two proxies - wing loading and specific lift - to reconstruct powered flight potential across the vaned feathered fossil pennaraptorans. The results recovered multiple origins of powered flight. We respectfully disagree with the criticism raised by Serrano and Chiappe2 that wing loading and specific lift, used in sequence, fail to discriminate between powered flight and gliding. We will explain this in reference to our original conservative approach.

Aerodynamics Show Membrane-Winged Theropods Were a Poor Gliding Dead-end.

The bizarre scansoriopterygid theropods Yi and Ambopteryx had skin stretched between elongate fingers that form a potential membranous wing. This wing is thought to have been used in aerial locomotion, but this has never been tested. Using laser-stimulated fluorescence imaging, we re-evaluate their anatomy and perform aerodynamic calculations covering flight potential, other wing-based behaviors, and gliding capabilities. We find that Yi and Ambopteryx were likely arboreal, highly unlikely to have any form of powered flight, and had significant deficiencies in flapping-based locomotion and limited gliding abilities. Our results show that Scansoriopterygidae are not models for the early evolution of bird flight, and their structurally distinct wings differed greatly from contemporaneous paravians, supporting multiple independent origins of flight. We propose that Scansoriopterygidae represents a unique but failed flight architecture of non-avialan theropods and that the evolutionary race to capture vertebrate aerial morphospace in the Middle to Late Jurassic was dynamic and complex.

Potential for Powered Flight Neared by Most Close Avialan Relatives, but Few Crossed Its Thresholds.

Uncertainties in the phylogeny of birds (Avialae) and their closest relatives have impeded deeper understanding of early theropod flight. To help address this, we produced an updated evolutionary hypothesis through an automated analysis of the Theropod Working Group (TWiG) coelurosaurian phylogenetic data matrix. Our larger, more resolved, and better-evaluated TWiG-based hypothesis supports the grouping of dromaeosaurids + troodontids (Deinonychosauria) as the sister taxon to birds (Paraves) and the recovery of Anchiornithinae as the earliest diverging birds. Although the phylogeny will continue developing, our current results provide a pertinent opportunity to evaluate what we know about early theropod flight. With our results and available data for vaned feathered pennaraptorans, we estimate the potential for powered flight among early birds and their closest relatives. We did this by using an ancestral state reconstruction analysis calculating maximum and minimum estimates of two proxies of powered flight potential-wing loading and specific lift. These results confirm powered flight potential in early birds but its rarity among the ancestors of the closest avialan relatives (select unenlagiine and microraptorine dromaeosaurids). For the first time, we find a broad range of these ancestors neared the wing loading and specific lift thresholds indicative of powered flight potential. This suggests there was greater experimentation with wing-assisted locomotion before theropod flight evolved than previously appreciated. This study adds invaluable support for multiple origins of powered flight potential in theropods (≥3 times), which we now know was from ancestors already nearing associated thresholds, and provides a framework for its further study. VIDEO ABSTRACT.

The fast and the frugal: Divergent locomotory strategies drive limb lengthening in theropod dinosaurs.

Limb length, cursoriality and speed have long been areas of significant interest in theropod paleobiology, since locomotory capacity, especially running ability, is critical in the pursuit of prey and to avoid becoming prey. The impact of allometry on running ability, and the limiting effect of large body size, are aspects that are traditionally overlooked. Since several different non-avian theropod lineages have each independently evolved body sizes greater than any known terrestrial carnivorous mammal, ~1000kg or more, the effect that such large mass has on movement ability and energetics is an area with significant implications for Mesozoic paleoecology. Here, using expansive datasets that incorporate several different metrics to estimate body size, limb length and running speed, we calculate the effects of allometry on running ability. We test traditional metrics used to evaluate cursoriality in non-avian theropods such as distal limb length, relative hindlimb length, and compare the energetic cost savings of relative hindlimb elongation between members of the Tyrannosauridae and more basal megacarnivores such as Allosauroidea or Ceratosauridae. We find that once the limiting effects of body size increase is incorporated there is no significant correlation to top speed between any of the commonly used metrics, including the newly suggested distal limb index (Tibia + Metatarsus/ Femur length). The data also shows a significant split between large and small bodied theropods in terms of maximizing running potential suggesting two distinct strategies for promoting limb elongation based on the organisms' size. For small and medium sized theropods increased leg length seems to correlate with a desire to increase top speed while amongst larger taxa it corresponds more closely to energetic efficiency and reducing foraging costs. We also find, using 3D volumetric mass estimates, that the Tyrannosauridae show significant cost of transport savings compared to more basal clades, indicating reduced energy expenditures during foraging and likely reduced need for hunting forays. This suggests that amongst theropods, hindlimb evolution was not dictated by one particular strategy. Amongst smaller bodied taxa the competing pressures of being both a predator and a prey item dominant while larger ones, freed from predation pressure, seek to maximize foraging ability. We also discuss the implications both for interactions amongst specific clades and Mesozoic paleobiology and paleoecological reconstructions as a whole.

Annotation of phenotypes using ontologies: a gold standard for the training and evaluation of natural language processing systems.

Natural language descriptions of organismal phenotypes, a principal object of study in biology, are abundant in the biological literature. Expressing these phenotypes as logical statements using ontologies would enable large-scale analysis on phenotypic information from diverse systems. However, considerable human effort is required to make these phenotype descriptions amenable to machine reasoning. Natural language processing tools have been developed to facilitate this task, and the training and evaluation of these tools depend on the availability of high quality, manually annotated gold standard data sets. We describe the development of an expert-curated gold standard data set of annotated phenotypes for evolutionary biology. The gold standard was developed for the curation of complex comparative phenotypes for the Phenoscape project. It was created by consensus among three curators and consists of entity-quality expressions of varying complexity. We use the gold standard to evaluate annotations created by human curators and those generated by the Semantic CharaParser tool. Using four annotation accuracy metrics that can account for any level of relationship between terms from two phenotype annotations, we found that machine-human consistency, or similarity, was significantly lower than inter-curator (human-human) consistency. Surprisingly, allowing curatorsaccess to external information did not significantly increase the similarity of their annotations to the gold standard or have a significant effect on inter-curator consistency. We found that the similarity of machine annotations to the gold standard increased after new relevant ontology terms had been added. Evaluation by the original authors of the character descriptions indicated that the gold standard annotations came closer to representing their intended meaning than did either the curator or machine annotations. These findings point toward ways to better design software to augment human curators and the use of the gold standard corpus will allow training and assessment of new tools to improve phenotype annotation accuracy at scale.

The wings before the bird: an evaluation of flapping-based locomotory hypotheses in bird antecedents.

BACKGROUND: Powered flight is implicated as a major driver for the success of birds. Here we examine the effectiveness of three hypothesized pathways for the evolution of the flight stroke, the forelimb motion that powers aerial locomotion, in a terrestrial setting across a range of stem and basal avians: flap running, Wing Assisted Incline Running (WAIR), and wing-assisted leaping. METHODS: Using biomechanical mathematical models based on known aerodynamic principals and in vivo experiments and ground truthed using extant avians we seek to test if an incipient flight stroke may have contributed sufficient force to permit flap running, WAIR, or leaping takeoff along the phylogenetic lineage from Coelurosauria to birds. RESULTS: None of these behaviours were found to meet the biomechanical threshold requirements before Paraves. Neither was there a continuous trend of refinement for any of these biomechanical performances across phylogeny nor a signal of universal applicability near the origin of birds. None of these flap-based locomotory models appear to have been a major influence on pre-flight character acquisition such as pennaceous feathers, suggesting non-locomotory behaviours, and less stringent locomotory behaviours such as balancing and braking, played a role in the evolution of the maniraptoran wing and nascent flight stroke. We find no support for widespread prevalence of WAIR in non-avian theropods, but can't reject its presence in large winged, small-bodied taxa like Microraptor and Archaeopteryx. DISCUSSION: Using our first principles approach we find that "near flight" locomotor behaviors are most sensitive to wing area, and that non-locomotory related selection regimes likely expanded wing area well before WAIR and other such behaviors were possible in derived avians. These results suggest that investigations of the drivers for wing expansion and feather elongation in theropods need not be intrinsically linked to locomotory adaptations, and this separation is critical for our understanding of the origin of powered flight and avian evolution.

Toward Synthesizing Our Knowledge of Morphology: Using Ontologies and Machine Reasoning to Extract Presence/Absence Evolutionary Phenotypes across Studies.

The reality of larger and larger molecular databases and the need to integrate data scalably have presented a major challenge for the use of phenotypic data. Morphology is currently primarily described in discrete publications, entrenched in noncomputer readable text, and requires enormous investments of time and resources to integrate across large numbers of taxa and studies. Here we present a new methodology, using ontology-based reasoning systems working with the Phenoscape Knowledgebase (KB; kb.phenoscape.org), to automatically integrate large amounts of evolutionary character state descriptions into a synthetic character matrix of neomorphic (presence/absence) data. Using the KB, which includes more than 55 studies of sarcopterygian taxa, we generated a synthetic supermatrix of 639 variable characters scored for 1051 taxa, resulting in over 145,000 populated cells. Of these characters, over 76% were made variable through the addition of inferred presence/absence states derived by machine reasoning over the formal semantics of the source ontologies. Inferred data reduced the missing data in the variable character-subset from 98.5% to 78.2%. Machine reasoning also enables the isolation of conflicts in the data, that is, cells where both presence and absence are indicated; reports regarding conflicting data provenance can be generated automatically. Further, reasoning enables quantification and new visualizations of the data, here for example, allowing identification of character space that has been undersampled across the fin-to-limb transition. The approach and methods demonstrated here to compute synthetic presence/absence supermatrices are applicable to any taxonomic and phenotypic slice across the tree of life, providing the data are semantically annotated. Because such data can also be linked to model organism genetics through computational scoring of phenotypic similarity, they open a rich set of future research questions into phenotype-to-genome relationships.

Moving the mountain: analysis of the effort required to transform comparative anatomy into computable anatomy.

The diverse phenotypes of living organisms have been described for centuries, and though they may be digitized, they are not readily available in a computable form. Using over 100 morphological studies, the Phenoscape project has demonstrated that by annotating characters with community ontology terms, links between novel species anatomy and the genes that may underlie them can be made. But given the enormity of the legacy literature, how can this largely unexploited wealth of descriptive data be rendered amenable to large-scale computation? To identify the bottlenecks, we quantified the time involved in the major aspects of phenotype curation as we annotated characters from the vertebrate phylogenetic systematics literature. This involves attaching fully computable logical expressions consisting of ontology terms to the descriptions in character-by-taxon matrices. The workflow consists of: (i) data preparation, (ii) phenotype annotation, (iii) ontology development and (iv) curation team discussions and software development feedback. Our results showed that the completion of this work required two person-years by a team of two post-docs, a lead data curator, and students. Manual data preparation required close to 13% of the effort. This part in particular could be reduced substantially with better community data practices, such as depositing fully populated matrices in public repositories. Phenotype annotation required ∼40% of the effort. We are working to make this more efficient with Natural Language Processing tools. Ontology development (40%), however, remains a highly manual task requiring domain (anatomical) expertise and use of specialized software. The large overhead required for data preparation and ontology development contributed to a low annotation rate of approximately two characters per hour, compared with 14 characters per hour when activity was restricted to character annotation. Unlocking the potential of the vast stores of morphological descriptions requires better tools for efficiently processing natural language, and better community practices towards a born-digital morphology. Database URL: http://kb.phenoscape.org

Finding our way through phenotypes.

Despite a large and multifaceted effort to understand the vast landscape of phenotypic data, their current form inhibits productive data analysis. The lack of a community-wide, consensus-based, human- and machine-interpretable language for describing phenotypes and their genomic and environmental contexts is perhaps the most pressing scientific bottleneck to integration across many key fields in biology, including genomics, systems biology, development, medicine, evolution, ecology, and systematics. Here we survey the current phenomics landscape, including data resources and handling, and the progress that has been made to accurately capture relevant data descriptions for phenotypes. We present an example of the kind of integration across domains that computable phenotypes would enable, and we call upon the broader biology community, publishers, and relevant funding agencies to support efforts to surmount today's data barriers and facilitate analytical reproducibility.

Annotation of phenotypic diversity: decoupling data curation and ontology curation using Phenex.

BACKGROUND: Phenex (http://phenex.phenoscape.org/) is a desktop application for semantically annotating the phenotypic character matrix datasets common in evolutionary biology. Since its initial publication, we have added new features that address several major bottlenecks in the efficiency of the phenotype curation process: allowing curators during the data curation phase to provisionally request terms that are not yet available from a relevant ontology; supporting quality control against annotation guidelines to reduce later manual review and revision; and enabling the sharing of files for collaboration among curators. RESULTS: We decoupled data annotation from ontology development by creating an Ontology Request Broker (ORB) within Phenex. Curators can use the ORB to request a provisional term for use in data annotation; the provisional term can be automatically replaced with a permanent identifier once the term is added to an ontology. We added a set of annotation consistency checks to prevent common curation errors, reducing the need for later correction. We facilitated collaborative editing by improving the reliability of Phenex when used with online folder sharing services, via file change monitoring and continual autosave. CONCLUSIONS: With the addition of these new features, and in particular the Ontology Request Broker, Phenex users have been able to focus more effectively on data annotation. Phenoscape curators using Phenex have reported a smoother annotation workflow, with much reduced interruptions from ontology maintenance and file management issues.

Body and limb size dissociation at the origin of birds: uncoupling allometric constraints across a macroevolutionary transition.

The origin of birds and powered flight is a classic major evolutionary transition. Research on their origin often focuses on the evolution of the wing with trends of forelimb elongation traced back through many nonavian maniraptoran dinosaurs. We present evidence that the relative forelimb elongation within avian antecedents is primarily due to allometry and is instead driven by a reduction in body size. Once body size is factored out, there is no trend of increasing forelimb length until the origin of birds. We report that early birds and nonavian theropods have significantly different scaling relationships within the forelimb and hindlimb skeleton. Ancestral forelimb and hindlimb allometric scaling to body size is rapidly decoupled at the origin of birds, when wings significantly elongate, by evolving a positive allometric relationship with body size from an ancestrally negative allometric pattern and legs significantly shorten by keeping a similar, near isometric relationship but with a reduced intercept. These results have implications for the evolution of powered flight and early diversification of birds. They suggest that their limb lengths first had to be dissociated from general body size scaling before expanding to the wide range of fore and hindlimb shapes and sizes present in today's birds.

Assessing arboreal adaptations of bird antecedents: testing the ecological setting of the origin of the avian flight stroke.

The origin of avian flight is a classic macroevolutionary transition with research spanning over a century. Two competing models explaining this locomotory transition have been discussed for decades: ground up versus trees down. Although it is impossible to directly test either of these theories, it is possible to test one of the requirements for the trees-down model, that of an arboreal paravian. We test for arboreality in non-avian theropods and early birds with comparisons to extant avian, mammalian, and reptilian scansors and climbers using a comprehensive set of morphological characters. Non-avian theropods, including the small, feathered deinonychosaurs, and Archaeopteryx, consistently and significantly cluster with fully terrestrial extant mammals and ground-based birds, such as ratites. Basal birds, more advanced than Archaeopteryx, cluster with extant perching ground-foraging birds. Evolutionary trends immediately prior to the origin of birds indicate skeletal adaptations opposite that expected for arboreal climbers. Results reject an arboreal capacity for the avian stem lineage, thus lending no support for the trees-down model. Support for a fully terrestrial ecology and origin of the avian flight stroke has broad implications for the origin of powered flight for this clade. A terrestrial origin for the avian flight stroke challenges the need for an intermediate gliding phase, presents the best resolved series of the evolution of vertebrate powered flight, and may differ fundamentally from the origin of bat and pterosaur flight, whose antecedents have been postulated to have been arboreal and gliding.