#PAGE_PARAMS# #ADS_HEAD_SCRIPTS# #MICRODATA#

A digital collection of rare and endangered lemurs and other primates from the Duke Lemur Center


Authors: Gabriel S. Yapuncich aff001;  Addison D. Kemp aff001;  Darbi M. Griffith aff001;  Justin T. Gladman aff003;  Erin Ehmke aff004;  Doug M. Boyer aff001
Authors place of work: Department of Evolutionary Anthropology, Duke University, Durham, North Carolina, United States of America aff001;  Department of Anthropology, University of Texas, Austin, Texas, United States of America aff002;  Shared Materials Instrumentation Facility (SMIF), Duke University, Durham, North Carolina, United States of America aff003;  Duke Lemur Center, Duke University, Durham, North Carolina, United States of America aff004
Published in the journal: PLoS ONE 14(11)
Category: Research Article
doi: https://doi.org/10.1371/journal.pone.0219411

Summary

Scientific study of lemurs, a group of primates found only on Madagascar, is crucial for understanding primate evolution. Unfortunately, lemurs are among the most endangered animals in the world, so there is a strong impetus to maximize as much scientific data as possible from available physical specimens. MicroCT scanning efforts at Duke University have resulted in scans of more than 100 strepsirrhine cadavers representing 18 species from the Duke Lemur Center. An error study of the microCT scanner recovered less than 0.3% error at multiple resolution levels. Scans include specimen overviews and focused, high-resolution selections of complex anatomical regions (e.g., cranium, hands, feet). Scans have been uploaded to MorphoSource, an online digital repository for 3D data. As captive (but free ranging) individuals, these specimens have a wealth of associated information that is largely unavailable for wild populations, including detailed life history data. This digital collection maximizes the information obtained from rare and endangered animals with minimal degradation of the original specimens.

Keywords:

Computed axial tomography – Primates – Iodine – Extremophiles – Feet – Lemurs – Bushbabies

Introduction

Lemurs, a group of primates endemic to Madagascar, are an important group of animals for understanding the evolutionary history and adaptive origins of primates. Unfortunately, they are among the most endangered mammals in the world, with 94% of lemur species threatened by extinction [1]. Continued habitat degradation and fragmentation, illegal poaching, and challenging economic and political circumstances in Madagascar mean that lemurs are likely to remain under acute threat in the foreseeable future [1]. While conservation groups have developed several local, site-specific action plans [2], protecting and studying these animals requires multiple strategies both in Madagascar and internationally.

The Duke Lemur Center (formerly the Duke University Primate Center) is a prime example of an alternative approach to the conservation and scientific study of lemurs. Founded in 1966 in Durham, North Carolina, the Duke Lemur Center (DLC) was established to operate as a “living laboratory” and permit non-invasive study of rare primates, including galagos, lorises, and lemurs (which together comprise the primate suborder Strepsirrhini). Over its history, the DLC has housed more than 4,000 individuals from 39 primate species, and currently houses nearly 240 individuals from 17 strepsirrhine species. The DLC is involved in multiple lemur conservation efforts, including 1) managing several breeding populations as an Association of Zoos and Aquariums accredited institution; 2) the SAVA conservation program, a community-based approach to sustainable forest management and economic improvement in northern Madagascar; and 3) working with the Malagasy government to develop animal husbandry, welfare, and breeding programs for ex-situ lemur populations in Madagascar.

While the cofounders of the DLC, Dr. John Buettner-Janusch and Dr. Peter Klopfer, ran research programs focused on genetics and behaviour respectively, the DLC’s unique resources have provided data for a wide variety of scientific fields, including anatomy and physiology [e.g., 36], social ecology [e.g., 78], cognition [e.g., 910, biomechanics [e.g., 1112], molecular biology [e.g., 1314], and palaeontology [e.g., 1518]. The importance and rarity of the animals housed at the DLC necessitates thorough and effective use in educational and research initiatives, and this spirit of efficiency extends to treatment of deceased individuals. When an animal dies at the DLC (most frequently of old age), veterinary staff perform necropsies to remove internal organs and then preserve cadavers in cold storage for future research purposes. There are currently more than 400 cadavers in storage. However, because DLC cadaveric specimens are available for destructive sampling for research purposes [e.g., 1921], the total information preserved by each specimen decreases over time. Digitizing the DLC’s cadaveric collection presents an opportunity to preserve hard tissue data without degrading the specimen’s anatomical integrity, thereby increasing the educational and scientific value of these rare and endangered animals.

Here we present an open access 3D digital collection of microCT scan data representing 113 adult individuals from the Duke Lemur Center. The data presented here were generated by GSY and ADK for use in their dissertations [22, 23]. All scans are publicly available on MorphoSource.org [24], an online repository specifically designed to archive 3D data. At the time of manuscript preparation, the collection consists of 483 TIFF volume stacks and 374 surface files generated from the volume data. The collection will continue to grow with future scanning efforts at Duke and potentially from contributions made by other researchers (in the form of new scans or surface files derived from the current collection). Primate cadavers have long been recognized as valuable scientific resources that could be utilized much more efficiently [25, 26], provided scientists motivated by different research questions could coordinate specimen access. The digital collection presented here follows several recent efforts to digitize and publish unique and valuable datasets [2729].

Materials and methods

DLC cadaveric collection

A total of 483 microCT (μCT) scans of strepsirrhine primates housed at the Duke Lemur Center were performed at the Shared Materials Instrumentation Facility at Duke University. Specimens represent both major clades of strepsirrhine primates, including 82 lemurs, 17 galagos, and 14 lorises (Fig 1). Among these individuals, two lemurs and five loris specimens were iodine-stained to permit visualization of soft tissue anatomy. Currently, the sample is biased toward individuals less than 3 kg. Individuals of larger species housed at the DLC (e.g., Propithecus and Varecia) have not yet been μCT scanned, although methods developed in the course of this project will allow them to be scanned at high resolutions in the near future. Additional demographic information for most specimens is available in Zehr et al. [30]. Table 1 provides a summary of the sample by species, and individual specimens, scanning parameters and digital object identifiers (DOIs) for all currently available TIFF volumes are provided in Table 2.

Strepsirrhine phylogeny and number of individuals included in this microCT collection.
Fig. 1. Strepsirrhine phylogeny and number of individuals included in this microCT collection.
Dark circles and bars indicate taxa currently represented in the collection. Open circles and bars represent taxa housed at DLC but not currently scanned. Chronogram downloaded from 10ktrees.org, version 3 [39].
Tab. 1. Summary of sample by species.
Summary of sample by species.
Tab. 2. Specimens, scan names, scan parameters, and DOIs for all TIFF stacks included in this collection.
Specimens, scan names, scan parameters, and DOIs for all TIFF stacks included in this collection.
Abbreviations: mm, millimeters; kV, kilovolts; uA, microamps; GB, gigabytes; MB, megabytes.

MicroCT scanning

A Nikon XT H 225 ST μCT machine at Duke University’s Shared Materials Instrumentation Facility (SMIF) was used for all scans. This machine has a Perkins Elmer AN1620 X-ray detector panel, which provides a 2000 x 2000 pixel field and a 7.5 frames per second readout. All scans were performed using a Nikon 225kV reflection target with a tungsten anode, which has a focal spot size ranging from 3 to 225 μm depending on the power. No filters were used. The voltage for these scans ranged between 75–203 kV and current settings ranged between 59–593 μA. These settings were largely dependent on the density of the specimen (higher density requires higher X-ray power to penetrate the specimen) and proximity of the detector panel to the reflection target (a smaller distance necessitates lower X-ray power to avoid oversaturation).

In order to ensure that each specimen could be used in subsequent research projects, we did not want scanning events to further degrade the specimen through thawing and refreezing. Therefore, we placed non-iodine stained specimens on dry ice in a Styrofoam cooler, stabilized the specimens with radiotransparent foam to prevent specimen movement during the scan, and conducted the scan with the specimens in the cooler. Iodine stained specimens were scanned in various containers and stabilized with radiotransparent foam but were not placed on dry ice. Variation in the resolution of the scans (voxel size) in Table 2 reflects the maximum level of magnification that could be achieved for an anatomical region given the dimensions of each specimen and the scanning container.

The protocol for this project prioritized comprehensiveness and detail. To that end, most specimens were scanned four to eight times and are represented by four to five volumes: an overview of the full body and separate higher resolution volumes of the skull, hands, and feet (Fig 2). Full body overviews are usually composites created by stitching together TIFF stacks from multiple scanning events (detailed further below).

Volume rendering of <i>Nycticebus coucang</i> (DLC 2901f) showing scanning protocol for strepsirrhine cadavers.
Fig. 2. Volume rendering of Nycticebus coucang (DLC 2901f) showing scanning protocol for strepsirrhine cadavers.
Boxes outline regions of increased anatomical complexity that were scanned at higher resolutions separately. To reduce noise, the threshold for grey values is lower than optimal threshold, rendering less dense bone transparent.

Iodine staining

The soft tissue anatomy of two lemur and five loris specimens (Table 1) was visualized in the CT-scans using iodine as a contrasting agent [31]. Fig 3 provides an example image from an individual scanned after the staining process. These specimens were thawed and fixed in 10% formalin (Carolina Biological) to prevent deterioration during staining. Specimens were stained in a 7% solution of Lugol’s Iodine (Carolina Biological) for six weeks prior to scanning to allow iodine to penetrate the tissues. Given the large volume of solution required to stain seven cadavers, Lugol’s solution was selected as the contrast-enhancing agent (rather than iosmium-tetroxide [3234] or phospho-molybdic acid [35]) due to its low toxicity, ease of access, ability to differentially stain types of soft tissue, and cost effectiveness. Following scanning, iodine was removed from the specimens by soaking them in a series of water baths over several weeks to leach the iodine [31] from the specimen’s tissues.

Volume rendering of iodine-stained <i>Cheirogaleus medius</i> (DLC 1657m) using Avizo software.
Fig. 3. Volume rendering of iodine-stained Cheirogaleus medius (DLC 1657m) using Avizo software.
A clipping plane in the software digitally slices through the fur and skin to show stained tissues underneath.

Post-processing and stitching

X-ray projections were reconstructed as 3D volumes using Nikon XTEK CT Pro 3D version XT 5.3.2, proprietary software purchased with the μCT machine and available to all μCT users at SMIF. Volumes were saved as 16-bit Tagged Image File Format (.TIFF) stacks. High resolutions scans were compressed with 7-Zip software and uploaded to MorphoSource.

If the size of the specimen prevented the full body overview from being done in a single scan, the overview was created by conducting a series of overlapping scans with a shared centre of rotation. This process was easy to accomplish when the specimen was in an extended posture and each scan was oriented along the vertical y-axis (Fig 4A), provided the specimen was not larger than the vertical travel distance permitted by the scanner’s dimensions. However, larger specimens were often flexed into C-shapes that exceeded the dimensions of the detector, even at the lowest magnification. In these cases, scans were conducted as a series of overlapping “panels” with two separate vertical axes (Fig 4B).

Examples of stitched composite scans.
Fig. 4. Examples of stitched composite scans.
A) composite scan of Microcebus murinus (DLC 893m) stitched along a single vertical axis; B) composite scan of Daubentonia madagascariensis (DLC 6604m) stitched along two vertical axes. Boxes indicate separate microCT scans, grey areas boxes indicate areas of scan overlap.

To generate full body composites, overlapping scans were stitched together in ImageJ [36] using the 3D Stitching plug-in [37]. As composites of two or more scanning events, the full body overviews are very large volumes. When composite overviews were extremely large (rendering subsequent processing too computationally demanding), we chose to upload the overlapping scans separately. If there are elements partially represented across two scans, these scans can be stitched together by researchers after download.

MicroCT error study

As discussed by Copes et al. [28], μCT scanners in academic instrumentation facilities accommodate a wide range of users with varying demands for scanning parameters (i.e., detector and stage settings, target type, beam settings). The flexibility required of μCT scanners in academic contexts stands in contrast to industrial or metrology-specific machines, which can be calibrated to maintain a degree of minimum error within a particular set of scan parameters. For scanners in academic instrumentation facilities, accuracy is determined by the initial installation settings and subsequent maintenance, with the assumption that measurement error is around 1%.

Here we expand the calibration study of Copes et al. [28] for SMIF’s Nikon XT H 225 ST μCT machine. To determine the accuracy of the scanner, three different standard spheres of known diameters (3.175mm, 6.35mm, and 12.7mm; machined with a +/- 1.0 μm tolerance) were scanned at a range of voxel resolutions. The 3.175mm and 6.35mm spheres was scanned at 5, 6, 7, 8, 9, 10, 15, and 20 μm per voxel with the detector fixed at its farthest position in the chamber. The 12.7mm sphere was scanned at 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100 μm per voxel with the detector in the same position. Each scan was collected at 175kV, 86μA (15W), 354ms, 2000 projections, 1 frame per projection, and without a filter. Nikon’s proprietary CT Pro 3D and CT Agent software reconstructed the projection data into a volume data file, which was then opened in VG Studio Max 2.2.

In VG Studio, an automatic surface determination was applied to the spheres, >20 fit points were placed on the surface, and an idealized sphere was fit to these points. Diameters produced from this measurement were recorded and compared to the reported diameter of the spheres. The relative percentage error (RE%) was calculated as the difference between the measured (MD) and reported diameters (RD), divided by the measured diameter (RE% = (MD-RD)/MD*100). Given that we are evaluating the same μCT machine, we expect relative % error to be similar to the <0.2% reported by Copes et al. [28].

Results

MicroCT error study

Table 3 reports error values for all three standard spheres, and error is plotted against voxel resolution in Fig 5. For each calibration sphere, we found less than 0.3% error at all resolution levels, with most scans demonstrating error levels below 0.1%.

Relative error at different voxel resolutions for three calibration spheres.
Fig. 5. Relative error at different voxel resolutions for three calibration spheres.
Spheres of known diameters of 3.175 mm, 6.35 mm, and 12.7 mm and were tested in the Nikon XT H 225 ST μCT scanner at Duke University’s Shared Materials Instrumentation Facility.
Tab. 3. Relative error at different voxel resolutions for three calibration spheres in the Nikon X-Tek XHT 225 ST scanner at Duke University’s Shared Materials and Instrumentation Facility.
Relative error at different voxel resolutions for three calibration spheres in the Nikon X-Tek XHT 225 ST scanner at Duke University’s Shared Materials and Instrumentation Facility.

Data records and availability

The μCT scans and associated metadata (Table 2) presented in this project are available through MorphoSource (http://www.morphosource.org), a free-to-use online repository designed to accommodate 3D data and its derivatives, including 3D surface renderings and other digital imagery. MorphoSource was created to address increasing demand for deposition and archiving of 3D digital data representing physical objects, and provides the necessary infrastructure to host, share, and manage 3D data for several different user types—from individual researchers to museum curators—allowing data authors and institutions to benefit from subsequent data use by third parties. Network storage is currently provided by Duke University’s data infrastructure with servers housed in multiple locations. Morphosource is currently supported by Duke University and the National Science Foundation.

The data files of the current project are available open access. Data files are copyrighted by the DLC and can be downloaded and re-used for non-commercial purposes (Creative Commons copyright license CC-BY NC). At the time of manuscript revision, the project had been viewed more than 58,000 times and more than 1900 data files have been downloaded. The most popular downloads have been TIFF volume stacks of Daubentonia madagascarensis, a unique and charismatic species not often found in museum collections. Because scanning efforts of the DLC cadaver collection are ongoing, this project will continue to grow to include new specimens and new species.

Discussion

Structure of the digital collection

We chose MorphoSource to host this collection because it offers several utilities for data authors and institutions. First, MorphoSource is structured to function as both an online platform for collaborative research and as a public-facing repository for 3D digital data, with high ease-of-use when transitioning between these two roles. For this project, we began to upload scans in 2016 and, for the next two years, data were shared privately with collaborators at multiple institutions. When the project matured, data were made searchable and downloadable to all MorphoSource users, signalling the project’s transition to a publicly facing data collection. In providing this dual functionality, MorphoSource has utility for data authors during all stages of 3D data collection.

Second, MorphoSource provides data authors and institutions multiple ways to track and summarize data usage, including summary reports of data usage as well as the ability to establish citable digital object identifiers (DOIs) for individual scans and derivative data. These tools provide metrics of the value of the collection and help provide professional benefits for researchers undertaking time-intensive data collection efforts. Table 2 was generated from the summary report of this project.

Third, while data collections can be made publicly available through MorphoSource, the platform allows data authors or institutions the option of managing user access through a data request system. This functionality, along with the ability to clearly delineate copyright restrictions and other terms of use, reduces concerns that data will be used in ways that are inconsistent with institutional goals.

Data files on MorphoSource are organized within a three-level hierarchy: specimens > media groups > media files. A fourth organizational unit, the MorphoSource project, exists at the same hierarchical level as the specimen (i.e., specimens are not contained within projects, although media records of a specimen might be). Projects are a collection of media records determined by the user and serve several purposes, including sharing multiple data files with collaborators easily, developing special collections of high public interest (e.g., K-12 Anthropology Teaching Collection, MorphoSource Project P158) or curating institutional holdings (such as this project).

On MorphoSource, specimens are digital representations of physical specimens stored in museums or other collections. As researchers upload new data, they increase the comprehensive sampling of MorphoSource specimens. When new specimen records are created, they can be linked with metadata provided by the home collection of the specimen through iDigBio (http://www.idigbio.org). In the case of this collection, metadata are maintained directly by project members, including DLC staff, as the DLC does not currently serve collection information to data aggregators or use interoperable metadata protocols.

Media groups are nested within specimens and generally represent discrete scanning or other imaging events. Metadata of the imaging event (including scan parameters, calibrations, and funding sources) are attached to media groups.

Media files are nested within media groups and represent raw data (such as TIFF stacks) and derivative data (such as surface renderings). When media files are made publicly available or “published”, they can be assigned DOIs, which function as permanent and direct links to the data as well as references for data citation in subsequent studies. Currently, the DLC cadaver project contains media files tagged by 1329 digital object identifiers. Further detail on the structural organization and associated metadata of MorphoSource can be found in Boyer et al. [24].

Computer requirements

By current computer standards, these μCT scans are large files. TIFF stacks can range from 1 to 13 GBs and the stitched full body composites are particularly large. System performance depends primarily on four hardware components: the graphics processing unit (GPU), the central processing unit (CPU), the amount of random-access memory (RAM), and the hard drive. For direct volume renderings or 3D surface visualization, the system should have a high-end graphics card (minimally a DDR5 memory interface with more than 2 GB RAM). Sufficient RAM is critical for 3D visualization and analyses; we follow Copes et al. (2016) in recommending computers have RAM that is at least twice as large as the largest TIFF stack. Image processing speed relies on the CPU, so processors with high clock speeds (greater than 3 gigahertz) are recommended. CPU clock speed is particularly important if volume rendering software is unable to access multiple cores. Finally, for uploading datasets quickly, solid state drives (SSD) provide much faster reading and writing speeds than traditional hard disk drives (HDD).

Other recommendations to improve performance for analysing 3D data are a 64-bit operating system and multiple processing cores (provided software compatibility). Specific 3D visualization programs (listed below) may have their own set of system recommendations, so researchers are encouraged to evaluate the match between their preferred software and available workstation.

Data manipulation

When downloaded from MorphoSource, these μCT scans will first need to be decompressed using open-source programs such as 7-Zip. After decompression, we recommend that users then open TIFF stacks with 2D image processing software such as Fiji [38] or ImageJ [36], as these programs permit some volume editing but require less memory than industrial volume visualization and analysis software. In Fiji or ImageJ, the user can easily adjust grey values to enhance contrast and crop the original μCT volume to regions of interest. Edited TIFF stacks can be saved as new image sequences that require less memory to open and manipulate. For 3D visualization and analyses, there are several commercially available programs (e.g., Avizo, Amira, Dragonfly, Mimix, Osirix, Spiers, and VG Studio Max) as well as freeware (e.g., Slicer3D).

Citation of scans

Usage terms are outlined in the MorphoSource user agreement that accompanies each download from the website. Subsequent publications that make use of these scans should include 1) a citation of this study; 2) a list or table of DOIs for each scan used; and 3) a statement of access accompanying the DOI list or in the acknowledgments. This statement should read “The Duke Lemur Center provided access to these data under a reuse but non-commercial creative commons license (CC BY-NC), originally appearing in Yapuncich et al. (2019), the collection and archiving of which was funded by NSF BCS 1540421 and NSF BCS 1552848. The files were downloaded from www.MorphoSource.org, Duke University.” A similar statement is included in the accompanying scan metadata.

Although not required by the usage agreement, we would urge researchers who conduct additional processing of these scans (e.g., generate surface files of elements of interest) to upload derivative files to MorphoSource. While we recognize that uploading derivatives can be a time-intensive process, we feel it is a critical component of data sharing and subsequent research. To facilitate this process, bulk upload options have been developed and are available through MorphoSource. Finally, we encourage researchers to contact the DLC research manager to obtain a DLC publication number when manuscripts using these data have been accepted for publication.


Zdroje

1. Schwitzer C, Mittermeier RA, Johnson SE, Donati G, Irwin M, Peacock H, et al. Averting lemur extinctions amid Madagascar's political crisis. Science. 2014; 343:842–843. doi: 10.1126/science.1245783 24558147

2. Schwitzer C, Mittermeier RA, Davies N, Johnson S, Ratsimbazafy J, Razafindramanana J, et al. Lemurs of Madagascar: A strategy for their conservation 2013–2016. Bristol, UK: IUCN SSC Primate Specialist Group, Bristol Conservation and Science Foundation, and Conservation International; 2013.

3. Cartmill M. Pads and claws in arboreal locomotion. In: Jenkins FA Jr, editor. Primate Locomotion. New York: Academic Press; 1974. pp. 45–83.

4. Kay RF. The functional adaptations of primate molar teeth. Am J Phys Anthropol. 1975; 43:195–215. doi: 10.1002/ajpa.1330430207 810034

5. Nishimura AC, Russo GA. Does cortical bone thickness in the last sacral vertebra differ among tail types in primates? Am J Phys Anthropol. 2017; 162:757–67. doi: 10.1002/ajpa.23167 28075029

6. Bornbusch SL, Greene LK, McKenney EA, Volkoff SJ, Midani FS, Joseph G, et al. A comparative study of gut microbiomes in captive nocturnal strepsirrhines. Am J Primatol. 2019; 13:e22986.

7. Klopfer PH, Jolly A. The stability of territorial boundaries in a lemur troop. Folia Primatol. 1970; 12:199–208. doi: 10.1159/000155289 5417920

8. Teichroeb JA, Vining AQ. Navigation strategies in three nocturnal lemur species: diet predicts heuristic use and degree of exploratory behavior. Anim Cognit. 2019; 22:343–354.

9. MacLean EL, Merritt DJ, Brannon EM. Social complexity predicts transitive reasoning in prosimian primates. Anim Behav. 2008; 76:479–486. doi: 10.1016/j.anbehav.2008.01.025 19649139

10. Samson DR, Vining A, Nunn CL. Sleep influences cognitive performance in lemurs. Anim Cog. 2019; 22:697–706.

11. Hylander WL. In vivo bone strain in the mandible of Galago crassicaudatus. Am J Phys Anthropol. 1977; 46:309–326. doi: 10.1002/ajpa.1330460212 403774

12. Fabre AC, Granatosky MC, Hanna JB, Schmitt D. Do forelimb shape and peak forces co‐vary in strepsirrhines? Am J Phys Anthropol. 2018; 167:602–614. doi: 10.1002/ajpa.23688 30159895

13. Buettner-Janusch J, Hill RL. Molecules and monkeys. Science. 1965; 147:836–842. doi: 10.1126/science.147.3660.836 14245766

14. Larsen PA, Harris RA, Liu Y, Murali SC, Campbell CR, Brown AD, et al. Hybrid de novo genome assembly and centromere characterization of the gray mouse lemur (Microcebus murinus). BMC Biol. 2017; 15:110. doi: 10.1186/s12915-017-0439-6 29145861

15. Cartmill M. Rethinking primate origins. Science. 1974; 184:436–443. doi: 10.1126/science.184.4135.436 4819676

16. Tattersall I. Cranial anatomy of the Archaeolemurinae (Lemuroidea, Primates). Anthropol Pap Am Nat Hist Mus. 1973; 1:1–110.

17. Gunnell GF, Boyer DM, Friscia AR, Heritage S, Manthi FK, Miller ER, et al. Fossil lemurs from Egypt and Kenya suggest an African origin for Madagascar’s aye-aye. Nature Commun. 2018; 9:3193.

18. Yapuncich GS, Feng HJ, Dunn RH, Seiffert ER, Boyer DM. Vertical support use and primate origins. Sci Reports. 2019; 9:12341.

19. Smith TD, Martell MC, Rossie JB, Bonar CJ, Deleon VB. Ontogeny and microanatomy of the nasal turbinals in lemuriformes. Anat Rec. 2016; 299:1492–510.

20. Huq E, Taylor AB, Su Z, Wall CE. Fiber type composition of epaxial muscles is geared toward facilitating rapid spinal extension in the leaper Galago senegalensis. Am J Phys Anthropol. 2018; 166:95–106. doi: 10.1002/ajpa.23405 29318571

21. Burrows AM, Omstead KM, Deutsch AR, Gladman JT, Hartstone-Rose A. Reverse dissection and diceCT reveal otherwise hidden data in the evolution of the primate face. JoVE: J Vis Exp. 2019; 143:e58394.

22. Yapuncich GS. Body mass prediction from dental and postcranial measurements in primates and their nearest relatives. Ph. D. Dissertation, Duke University. 2017.

23. Kemp AD. Primate binocular vision: adaptive significance in grasping and locomotion. Ph. D. Dissertation, University of Texas, Austin. 2019.

24. Boyer DM, Gunnell GF, Kaufman S, McGeary TM. Morphosource: Archiving and sharing 3-d digital specimen data. Paleontol Soc Pap. 2016; 22:157–181.

25. Mittermeier RA, Fleagle JG. A primate distribution program to end wastage of sacrificed specimens. Lab Prim News. 1973; 12: 1–3.

26. Gordon AD, Marcus E, Wood B. Great ape skeletal collections: making the most of scarce and irreplaceable resources in the digital age. Am J Phys Anthropol. 2013; 152:2–32. doi: 10.1002/ajpa.22391 24249590

27. Adams JW, Olah A, McCurry MR, Potze S. Surface model and tomographic archive of fossil primate and other mammal holotype and paratype specimens of the Ditsong National Museum of Natural History, Pretoria, South Africa. PLoS One. 2015; 10:e0139800. doi: 10.1371/journal.pone.0139800 26441324

28. Copes LE, Lucas LM, Thostenson JO, Hoekstra HE, Boyer DM. A collection of non-human primate computed tomography scans housed in MorphoSource, a repository for 3D data. Sci Data. 2016; 3:160001. doi: 10.1038/sdata.2016.1 26836025

29. Shi JJ, Westeen EP, Rabosky DL. Digitizing extant bat diversity: An open-access repository of 3D μCT-scanned skulls for research and education. PLoS One. 2018; 13:e0203022. doi: 10.1371/journal.pone.0203022 30226875

30. Zehr SM, Roach RG, Haring D, Taylor J, Cameron FH, Yoder AD. Life history profiles for 27 strepsirrhine primate taxa generated using captive data from the Duke Lemur Center. Sci Data. 2014; 1:140019. doi: 10.1038/sdata.2014.19 25977776

31. Gignac PM, Kley NJ, Clarke JA, Colbert MW, Morhardt AC, Cerio D, et al. Diffusible iodine‐based contrast‐enhanced computed tomography (diceCT): an emerging tool for rapid, high‐resolution, 3‐D imaging of metazoan soft tissues. J Anat. 2016; 228:889–909. doi: 10.1111/joa.12449 26970556

32. Metscher BD. Micro-CT for comparative morphology: simple staining methods allow high-contrast 3D imaging of diverse non-mineralized animal tissues. BMC Physiol. 2009a; 9:11.

33. Metscher BD. Micro-CT for developmental biology: a versatile tool for high-contrast 3-D imaging at histological resolutions. Dev Dyn. 2009b; 238:632–640.

34. Mizutani R, Suzuki Y. X-ray microtomography in biology. Micron 2012; 43:104–115. doi: 10.1016/j.micron.2011.10.002 22036251

35. Pauwels E, Van Loo E, Cornillie P, Brabant L, van Hoorebeke L. An exploratory study of contrast agents for soft tissue visualization by means of high resolution X-ray computed tomography imaging. J Microsc. 2013; 250:21–31. doi: 10.1111/jmi.12013 23432572

36. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nature Methods. 2012; 9:671. doi: 10.1038/nmeth.2089 22930834

37. Preibisch S, Saalfeld S, Tomancak P. Globally optimal stitching of tiled 3D microscopic image acquisitions. Bioinformatics. 2009; 25:1463–1465. doi: 10.1093/bioinformatics/btp184 19346324

38. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological-image analysis. Nature Methods. 2012; 9:676. doi: 10.1038/nmeth.2019 22743772

39. Arnold C, Matthews LJ, Nunn CL. The 10kTrees website: a new online resource for primate phylogeny. Evol Anthropol. 2010; 19:114–118.


Článek vyšel v časopise

PLOS One


2019 Číslo 11
Nejčtenější tento týden
Nejčtenější v tomto čísle
Kurzy Podcasty Doporučená témata Časopisy
Přihlášení
Zapomenuté heslo

Zadejte e-mailovou adresu, se kterou jste vytvářel(a) účet, budou Vám na ni zaslány informace k nastavení nového hesla.

Přihlášení

Nemáte účet?  Registrujte se

#ADS_BOTTOM_SCRIPTS#