Table of Contents

• Dataset Description
  • Methods & Sampling
  • Data Processing Description
• Data Files
• Parameters
• Instruments
• Deployments
• Project Information
• Funding

This data set contains Raman spectra of the following species: Argopecten irradians, Crassostrea gigas, Crassostrea virginica, Gemma gemma, Geukensia demissa, Ischadium recurvum, Macoma mitchelli, Mercenaria mercenaria, Mulinia lateralis, Mya arenaria, Mytilopsis leucophaeata, Ostrea lurida, Rangia cuneata, Tagelus plebeius, Spisula solidissma, Panopea generosa. Detailed information on the source and preparation of larvae can be found in Thompson et al. (in prep). All spectra were acquired by Adam Schlenger and Christine Thompson.

Detailed information on the source and preparation of larvae can be found in Thompson et al. (in prep). All spectra were acquired by Adam Schlenger and Christine Thompson. The following information on the acquisition and processing of the spectra that comprise this dataset was quoted from Thompson et al. (in prep):

"Raman spectra were acquired with an XploRA confocal Raman microscope by Horiba Jobin Yvon, Inc. The system includes a flat field spectrograph with a multichannel air cooled CCD detector and color camera optically coupled to an Olympus BX41 microscope. We used three lasers: a 532 nm 25 mW solid-state laser, a 638 nm 25 mW laser diode, and a 785 nm 25 mW laser diode. The lasers ran through a 100x objective using a 1200 grove mm-1 grating and hole and slit size of 300 µm and 100 µm, respectively. Spectra were recorded in the range of 200-2000 cm-1. Spectra were acquired from 20 larval shells for each sample by averaging 3 accumulations with an exposure time of 10 s. For one shell in each sample, three spectra were taken from different positions on the shell. Spectral acquisition was controlled using Horiba’s LabSpec software (version 6). Wavelength calibration was performed on the XploRA system using a neon light source that was calibrated daily with a silicon wafer.

All spectra were first pre-processed to remove noise and other variability. Immediately after acquisition, noise was removed using a smoothing function in LabSpec. Baseline correction was then performed using a freely-available integrated software system for processing Raman spectra (Reisner et al. 2011) implemented in MATLAB (v. R2011a). Next, all wavenumbers were shifted to ensure the aragonite peak for all spectra fell at 1085 cm-1 and then spectra were standardized to the intensity of the aragonite peak on a scale of 0 to 1."

The two columns of numbers in each file correspond to wavenumber with units of cm-1 (first column) and relative intensity which is unitless (second column).

References:
Reisner, L.A., Cao, A., & Pandya, A.K. 2011. An integrated software system for processing, analyzing, and classifying Raman spectra. Chemom Intell Lab Syst 105:83–90. doi: 10.1016/j.chemolab.2010.09.011

Thompson, C.M., North, E.W., Kennedy, V.S., & White, S.N. In prep. Classifying bivalve larvae using shell pigments identified by Raman spectra. Analytical and Bioanalytical Chemistry.

BCO-DMO downloaded the original data files from the DropBox URL provided by Dr. E. North. BCO-DMO created additional data columns for serving based on the file name conventions and metadata. See the original metadata form describing the file name conventions: Metadata_for_Raman_Spectra_of_Bivalve_Larvae (PDF file).

Identification of bivalve larvae is notoriously difficult and time consuming; currently no ‘gold standard’ method exists for distinguishing larvae collected in the field. We proposed to determine if Raman spectroscopy could be used to identify species of bivalve larvae. Raman spectroscopy is a non-destructive method that uses a focused laser to produce a spectrum, a graph with peaks that indicate the presence of different molecules like calcium carbonate and organic pigments. Previous work exploring the use of Raman spectroscopy on shells of bivalve larvae revealed that larval shells of the eastern oyster (Crassostrea virginica) exhibit a unique spectrum at one wavelength which was distinguishable from those of six other bivalve species. The objective of this research was to analyze additional species of larvae and to determine if spectra collected at three different wavelengths could be used to distinguish the larvae. We collected Raman spectra at three wavelengths from 25 samples of bivalve larvae representing 16 species and four taxonomic Orders. Use of the Raman spectra with three wavelengths enabled classification of larvae into Order/Family groups with accuracies ≥ 92% (Thompson et al. in prep). This study indicates that continued evaluation of this approach would be a fruitful line of research which could advance knowledge of shellfish biology and ecology.

(Project description provided by Dr. Elizabeth North.)