For being able to identify minor spectral differences due to subtle changes in molecular structure, high sensitivity is key—particularly for tertiary structure assessment by near-UV CD and HOS comparisons, e.g., for forced degradation studies.
THE WALK-AWAY CIRCULAR DICHROISM INSTRUMENT FOR REGULATORY ENVIRONMENTS
Chirascan Q100 provides detailed insight into the Higher Order Structure (HOS) characteristics of complex biomolecules.
Reproducible robotics and high performance circular dichroism spectroscopy combine to generate quality data compatible with the most stringent statistical analysis methods. The result: objective, statistically-validated HOS comparisons.
Fully integrated for unattended operation, Chirascan Q100 saves days of operator time, analyzing 48 buffer-sample pairs over 24 hours whereas an experienced operator processes up to 14 samples per day on a manual system.
Performance
Increased sensitivity for superior detection
- Avalanche photodiode detector enhances sensitivity
- Higher signal:noise ratio than conventional photomultiplier tube detectors
- Accurate normalization from simultaneous measurement of absorbance and CD
Calibration with confidence
- Accurate CD values across entire wavelength range
- Overcome challenges of chemical calibration
- Optics-based, multiwavelength calibration
Conventional chemical calibration methods require considerable skill in preparation. Standards, such as camphor-10-sulfonic acid (CSA), are unstable, photolabile, and hygroscopic. In addition, single-wavelength calibration (290.5 nm) assumes the same linear response at all wavelengths.
The optics-based, multiwavelength calibration method used for Chirascan V100 and Chirascan Q100 systems overcomes these challenges. The correct calibration is applied to every wavelength to yield accurate CD values.
Publications
Features and accessories
Chirascan systems are supplied with the features and accessories required for acquisition of high quality data from ‘day one’.
- Removes common gas impurities
- Regulates purge gas consumption
- Software-controlled
- Software-controlled
- Up-time recorded
- Produces horizontally, linearly polarized monochromatic light
- Dual polarizing prisms maximize light throughput
- Converts horizontally polarized light to circularly polarized light. Alternates between left- and right-handed circular polarized light
- Eliminate sample handling errors
- Precise liquid handling and reproducibility
- Temperature-controlled storage maintains sample integrity
- Consistent analytical conditions
- Continuous temperature ramps (single sample mode)
- Dissipates heat from sample chamber and sample storage Peltiers
- Highest sensitivity (high signal: noise)
- For CD accuracy at every wavelength
- Eliminates errors of cuvette handling
- Recognized by Chirascan Control to select run/wash/dry protocols
- Choice of pathlength to optimize concentration and absorbance
- Selected for far- and near-UV CD analysis of biomolecules (single sample mode)
Customer Experiences
Dr Nick Larson
Dr. Nick Larson, a seasoned expert in protein and oligonucleotide therapeutics, leverages the Chirascan CD spectrometer in his work on a wide range of therapeutics, from protein-based drugs to advanced gene-silencing molecules.
HOS Comparison Software
Optimized for Chirascan Q100
- Confirming the statistical significance of any detected change in HOS requires a minimum of five sample replicates to enable robust analysis*. HOS comparison software is therefore optimized for analysis of the raw (unsmoothed) data generated on a Chirascan Q100 system to ensure
- An ability to detect minor changes in HOS
- An autosampler and flow cell combine to increase reproducibility by eliminating variability due to manual handling of samples and cuvettes
- An avalanche photodiode detector achieves a sensitivity far greater than a conventional photomultiplier detector
- An ability to increase number of replicates without increasing demands on operator time
The software includes comparison methods and statistics associated with several statistical methods. However, we recommend calculating z-scores to quantify the similarity between a single spectrum and the reference dataset for each sample using ‘weighted spectral difference’ Dinh et al.
For more information about quality range tests consult guidance documents from the relevant regulatory agencies. For example: Development of Therapeutic Protein Biosimilars: Comparative Analytical Assessment and Other Quality-Related Considerations – Guidance for Industry, CDER, FDA
*Analyses with three or four replicates can be sufficient for proof of principle studies, but statistical analysis with less than three replicates is not mathematically possible
Note: t-tests are not recommended for HOS comparisons and are absent from the statistical approaches recommended by regulatory agencies at this time.
- HOS Comparison software enables calculation of weighted spectral difference scores for reference and sample datasets – converting spectra into numerical data.
- These scores are then exported and used to generate a quality range test or equivalence test as recommended by the FDA.