Orchid fruit and root movement analyzed using 2D photographs and a bioinformatics pipeline for processing sequential 3D scans

Dewi Pramanık, Lotta M. Vaskimo, Kees Joost Batenburg, Alexander Kostenko, Kevin Droppert, Erik Smets, Barbara GravendeelApplications 2024 · DOI

TL;DR

This paper introduces a novel bioinformatics pipeline that combines 3D micro-computed tomography (CT) and 2D photography to precisely track the complex movements of orchid roots and fruits during development. This approach provides unprecedented spatial and temporal detail, crucial for understanding the genetic mechanisms behind these movements and for future transcriptomic studies.

What Problem Does This Paper Address?

Understanding plant organ movement, particularly twisting and resupination (a 180° rotation), is vital for plant development and survival. While resupination is a hallmark of orchids, most studies rely on basic morphological observations. This lack of precise, quantitative data makes it difficult to pinpoint the exact timing and spatial dynamics of these movements, which is essential for identifying the underlying genetic and molecular mechanisms. Current imaging techniques, like time-lapse photography, often lack the detail or can result in organs moving out of view, limiting their usefulness for detailed analysis and for harvesting specific tissues at precise developmental stages for molecular studies.

Methods

The researchers developed an integrated bioinformatics pipeline to process sequential 3D micro-CT scans and 2D photographs.

  • 3D CT Scanning: For fast-growing Erycina pusilla roots, high-resolution 3D micro-CT scans were performed weekly over several weeks. This technique captures detailed volumetric data of the root structure and its changes over time.
  • Bioinformatics Pipeline: A custom pipeline was built using Python and existing open-source software (FleXbox, 3D Slicer, FFmpeg). This pipeline automates the processing of raw CT data into 3D time-lapse videos. Key steps include:
  • Data Pre-processing (rawedit.py): Standardizing scan data, especially when the entire sample doesn't fit in a single scan, by trimming images to consistent volumes.
  • Image Alignment and Binning (timelapce.py): Registering and aligning 3D reconstructions to ensure smooth temporal transitions and reducing data size through binning.
  • 3D Visualization and Recording (slicer.py): Using 3D Slicer to visualize the reconstructed volumes and record rotating views of the moving organs.
  • Video Compilation (FFmpeg): Stitching the recorded screenshots into a final 3D time-lapse video.
  • 2D Photography: For slower-developing E. pusilla and Phalaenopsis equestris fruits, weekly 2D photographs were taken using a standardized setup (copy stand, lighting) to ensure consistent image quality and positioning.

Datasets

The study utilized two orchid species:

  • Erycina pusilla: Both roots and fruits were analyzed. Roots were scanned using 3D CT, and fruits were documented with 2D photographs.
  • Phalaenopsis equestris: Fruits were documented using 2D photographs.

The paper also makes available the scripts for the 3D time-lapse pipeline on GitHub and the raw reconstruction data for young and mature E. pusilla plants on Figshare.

Key Insights and Ideas

  • Independent Movement: Each orchid root and fruit exhibits an independent direction and degree of torsion, meaning their movements are not synchronized or dictated by a single global process.
  • Precise Temporal Mapping: The pipeline allows for the precise identification of when resupination begins and ends, crucial for future genetic studies.
  • X-ray Impact: Repeated X-ray exposure from CT scanning can significantly slow down plant development, even delaying flowering. This highlights the need for careful experimental design, especially when studying long-term processes.
  • Orchid Resupination is Complex: The study reveals intricate patterns of twisting and resupination in both roots and fruits, occurring at different developmental stages and with varying degrees of rotation.

Key Results

  • Erycina pusilla roots: Showed multiple twisting and resupination events from early development (77-84 days after germination [DAG]) through later stages (140-154 DAG).
  • Erycina pusilla fruits: Twisted approximately 45° from 56–63 days after pollination (DAP).
  • Phalaenopsis equestris fruits: Began resupination about a week before dehiscence (133 DAP) and continued a week after (161 DAP), with movements ranging from 20° to 180°.
  • The developed bioinformatics pipeline successfully generated detailed 3D time-lapse videos of orchid root and fruit movement.

Limitations and Future Work

  • X-ray Damage: The primary limitation identified is the detrimental effect of repeated X-ray exposure on plant development, leading to delayed growth and flowering. The authors recommend using seedlings or immature plants, minimizing X-ray dose, and supplementing with non-invasive photography for initial observations.
  • Misalignment in 3D Reconstruction: For E. pusilla, fan-shaped leaf morphology caused some misalignments in 3D reconstructions, hindering precise tracking. Future work could involve using markers, fixed mounts, or developing new algorithms for alignment.
  • Transcriptomic Integration: The authors emphasize the need for transcriptomic analysis to identify genes responsible for resupination. They suggest harvesting root tissue at specific times (e.g., 70 DAG for E. pusilla seedlings, 147 DAG for mature roots) for optimal results.
  • Further Imaging Refinements: Suggestions include adding low concentrations of metal ions to the growth medium to improve scan resolution and exploring alternative imaging strategies to minimize X-ray exposure.

Summary updated

Abstract

Our methods revealed that each orchid root and fruit had an independent direction and degree of torsion from the initial to the final position. Our innovative approaches produced detailed spatial and temporal information on the resupination of roots and fruits during orchid development.
BibTeX
@inproceedings{pramank2024orchid,
  author = {Dewi Pramanık and Lotta M. Vaskimo and Kees Joost Batenburg and Alexander Kostenko and Kevin Droppert and Erik Smets and Barbara Gravendeel},
  title = {{Orchid fruit and root movement analyzed using 2D photographs and a bioinformatics pipeline for processing sequential 3D scans}},
  booktitle = {Applications},
  year = {2024},
  doi = {10.1002/aps3.11567},
}
Figure