Welcome

Join the hundreds of researchers from around the world who are part of the Open Specy community by analyzing, sharing, processing, and identifying their Raman and IR spectra.

on Twitter

Watch us develop Open Specy on GitHub, file an Issue, or request a feature

Or just e-mail [email protected] to be added to the Open Specy mailing list


Open Specy is free and open source thanks to our partners.


Looking for the classic version of OpenSpecy? Go to wincowger.shinyapps.io/openspecy-classic

  1. Upload a .zip, .csv, .0, .asp, .jdx, .spc, or .spa file to the upload file tab.
  2. Process your data using smoothing, derivative, baseline correction, flattening, range selection, and intensity adjustment.
  3. Identify your spectra using onboard reference libraries and/or AI
  4. Download your results
  5. For more details click the button below for the SOP or watch the detailed instructional videos.
SOP



Signal thresholding technique, value, and histogram threshold plot.

Min-Max normalization improves comparability, for many applications, between spectra except in cases where raw intensity values are necessary for interpreation. For example raw values can be useful for thresholding. Min-Max normalization rescales spectral intensity values between 0-1

Smoothing can enhance signal to noise and uses the SG filter with the polynomial order specified, 3 default usually works well. Derivative transformation uses the order specified. If doing identification with a derivative library, 1 is required, 0 should be used if no derivative transformation is desired. Smoothing uses the SG filter on an window of points, specifying the wavenumber window larger will make the spectra more smooth. The absolute value does something similar to intensity correction to make the spectra more absorbance-like.


Options for conforming spectra to a new wavenumber resolution. Conformation technique specifies the strategy for performing the conformation. Nearest will use the nearest value to the wavenumber resolution specified, this is faster but less accurate. Linear Interpolation will perform a linear regression between the nearest points to identify the intensity values at the new wavenumbers. Wavenumber Resolution will set the step size in wavenumbers for the new wavenumber values.

Open Specy assumes spectra are in Absorbance units. If the uploaded spectrum is not in absorbance units, use this input to specify the units to convert from.The transmittance adjustment uses the log10(1/T) calculation which does not correct for system and particle characteristics. The reflectance adjustment uses the Kubelka-Munk equation (1-R)2/(2*R). We assume that the reflectance is formatted as a percent from 1-100 and first correct the intensity by dividing by 100 so that it fits the form expected by the equation. If none is selected, Open Specy assumes that the uploaded data is an absorbance spectrum.

This algorithm automatically fits to the baseline by fitting polynomials of the provided order to the whole spectrum using the iModPolyFit algorithm.

Restricting the spectral range can remove regions of spectrum where no peaks exist and improve matching. These options control the maximum and minimum wavenumbers in the range to crop the spectra.

Sometimes peaks are undersireable. These options will replace peak regions with the mean of their edges. Specify the edge locations of the peaks minimum and maximum wavenumbers to use for flattening. Defaults are set to flatten the CO2 region in infrared spectra.
Options for processing the spectra.

These options define the strategy for identification. The ID Library will inform which library is used. Both (default) will search both FTIR and Raman libraries. Deriv will search against a derivative transformed library. No Baseline will search against a baseline corrected library. This should be in line with how you choose to process your spectra. Cor options use a simple Pearson correlation search algorithm. AI is uses either a multinomial model (experimental) or correlation on mediod (default) spectra from the library. Correlation thresholding will set the minimum value from matching to use as a 'positive identification'

Spectra

Upload some data to get started...

Help us reach our goal to revolutionize spectroscopy.


Monetary Partners

Thriving (10,000–100,000$)

Maintaining (1,000–10,000$)

Supporting (100–1,000$)

Jennifer Gadd

Saving (<100$)

Anne Jefferson, Heather Szafranski, Gwendolyn Lattin, Collin Weber, Gregory Gearhart, Anika Ballent, Shelly Moore, Susanne Brander (Oregon State University), Jeremy Conkle (TEXAS A&M UNIVERSITY CORPUS CHRISTI)

In-Kind Partners

Thriving (10,000–100,000$)

Win Cowger, Zacharias Steinmetz

Maintaining (1,000–10,000$)

Garth Covernton, Jamie Leonard, Shelly Moore, Rachel Kozloski, Katherine Lasdin, Aleksandra Karapetrova, Laura Markley, Walter Yu, Walter Waldman, Vesna Teofilovic, Monica Arienzo, Mary Fey Long Norris, Cristiane Vidal, Scott Coffin, Charles Moore, Aline Carvalho, Shreyas Patankar, Andrea Faltynkova, Sebastian Primpke, Andrew Gray, Chelsea Rochman, Orestis Herodotu, Hannah De Frond, Keenan Munno, Hannah Hapich, Jennifer Lynch

Supporting (100–1,000$)

Alexandre Dehaut, Gabriel Erni Cassola

To share spectra upload a file to the upload file tab. If you selected Share a copy of your spectra will be sent to the Community Data Warehouse on Open Science Framework. To add additional metadata, fill in the avaliable metadata fields and click -Share Data-. The spectra file that you uploaded along with your responses will be copied to the a -With Metadata- subfolder at the link below. All shared data holds a Creative Commons Attribution License 4.0.

We are a group of experienced spectroscopists and can provide a variety of services for hire, please contact [email protected] to inquire about any of the services below.

  • Adding new features to OpenSpecy
  • Creating spectroscopy software
  • Microplastic sample analysis
  • Spectral identification
  • Study design
  • So much more...