DATA

Actinotrocha larvae of Phoronis muelleri were fixed with 2% glutaraldehyde in 0.1M sodium cacodylate buffer at a pH value of 7.4. Subsequently, specimens were post-fixed using 1%OsO4. For microCT scanning, a specimen was stained with I2KI (0.1% (w/v) elemental iodine (I2) and 0.2% (w/v) potassium iodide (KI) in distilled water) and embedded in 1.5% low melt Agarose in a plastic pipette tip. For light microscopic investigation, a specimen was embedded in Agar low viscosity resin (Agar Scientific, Stansted, UK) and serially sectioned using a Jumbo diamond knife (Diatome AG, Biel, Switzerland) on a Reichert UltraCut-S microtome with a section thickness of 0.5µm.

Specimens of Oikopleura sp. were fixed with 2% glutaraldehyde in 0.1M sodium cacodylate buffer at a pH value of 7.4. Subsequently, specimens were post-fixed using 1%OsO4. For microCT scanning, a specimen was stained with I2KI (0.1% (w/v) elemental iodine (I2) and 0.2% (w/v) potassium iodide (KI) in distilled water) and embedded in 1.5% low melt Agarose in a plastic pipette tip. For light microscopic investigation, a specimen was embedded in Agar low viscosity resin (Agar Scientific, Stansted, UK) and serially sectioned using a Jumbo diamond knife (Diatome AG, Biel, Switzerland) on a Reichert UltraCut-S microtome with a section thickness of 0.5µm.

Specimens of Noctiluca scintillans were fixed with 4% buffered formalin. For microCT scanning, a specimen was stained with I2KI (0.1% (w/v) elemental iodine (I2) and 0.2% (w/v) potassium iodide (KI) in distilled water) and mounted in distilled water in a plastic pipette tip. For light microscopic investigation, a specimen was post-fixed using 1%OsO4 and embedded in Agar low viscosity resin (Agar Scientific, Stansted, UK) and serially sectioned using a Jumbo diamond knife (Diatome AG, Biel, Switzerland) on a Reichert UltraCut-S microtome with a section thickness of 0.5µm.

Specimens of Tomopteris helgolandica were fixed with 4% buffered formalin. For microCT scanning, a specimen was stained with I2KI (0.1% (w/v) elemental iodine (I2) and 0.2% (w/v) potassium iodide (KI) in distilled water) and mounted in distilled water in a plastic pipette tip.

Serial light microscopical sections were imaged using an Aperio ScanScope slide scanner with a 20x objective. Single sections were cropped from whole slide images in Adobe Photoshop CS5 (Adobe, San José, CA). These light microscopic images and videos were used as motion and color references.

The image volumes generated by microCT and from serial light microscopical sections were used for segmentation of the whole animal outline and specific features of each animal, such as the alimentary tract. Image segmentation was performed with the 3D software package Amira 6.1. (FEI Visualization Sciences Group, Mérignac Cédex, France) by combining intensity-based and manual segmentation tools. Based on the image segmentation, polygon surface meshes were triangulated. Subsequently the surfaces were smoothed and re-meshed in order to obtain isotropic surfaces triangles.

The 3D models acquired through image segmentation have to be edited further in order to make these animatable. The polygon resolution of the 3D model should be reduced. It is often a time-consuming task to achieve an animation-ready 3D model. The data editing must be carefully done to only reduce the information that is not necessary for accurate depictions. It is important to go back and forth with biological experts to not obscure critical details. This data editing process and related parameters depend on the desired end application of the 3D models. The animatable 3D model needs a material that defines the surfaces for the computation of the final images. For the animations, digital bones assist in moving the parts of the 3D models of the plankton organisms according to video references. The process of preparing and animating 3D models differs in complexity and depends greatly on the quality of the scanning data. Therefore, we present all the worksteps in a generalized workflow.