lab

 

After developing the metadata categories and items that characterize chemical analysis it was time to turn to the development of an ontology for chemical analysis.  Development begin by reviewing the entries and organization structure of the following ontologies

As a resullt the Chemical Analysis Ontology (CAO) is organized using the BFO framework with entries primarily under concepts (not complete by any means), material entities, information content entities (data items), roles, and processes. Currently, the ontology is more of a vocabulary in that development of predicates (ontology properties) that relate subjects to objects has not been a focal point. This will be fleshed out over the summer of 2015.

Graphical snapshots of the  0.1 version are show below and the current version of the ontology (0.2 - 4/29/15) in RDF/OWL (using OWL-DL) is linked here.  Please feed any comments to This email address is being protected from spambots. You need JavaScript enabled to view it..

Material Entities

material entities

Concepts (need a lot of work)

concepts

Information Content Entities

data items

Roles

expanded roles

One of the common ways that scientists currently annotate research papers for peer-review publication is to a keywords, or better yet, concept terms.  While it is impractical for ChAMP to define lists of technique specific concept terms it should i) provide a generic set of concept terms for chemical analysis and ii) provide a mechanism for inclusion of more specific concepts from existing or future lists developed by technique experts (e.g. IUPAC Orange Book).  Send any comments to This email address is being protected from spambots. You need JavaScript enabled to view it..

  • concept: a particular concept in the context of a chemical analysis that helps to clarify important aspects of the methodology [02-17-15]
  • scope: the extent of applicability (in terms of analysis technique or chemical analysis) of the concept [02-17-15]
  • vocabulary: reference to a defined controlled vocabulary/ontology [02-17-15]

To contextualize the chemical analysis data it is important to identify the people involved, roles they play, and the places that they analysis was developed.

  • contact: a specific individual that can be contacted about the chemical analysis (string) [2-12-15]
  • person: an individual that has participated in part in the development/production/publication of the chemical analysis (string) [1-19-15]
  • organization: a company/institution/organization that was part of the development/production/publication of the chemical analysis (string) [1-19-15]
  • funding agency: a public or private group that was a source of funding relative to the chemical analysis (string) [1-19-15]
  • role: the part that a contact plays in the development/production/publication of the chemical analysis (string/enum) [1-19-15]
  • address: physical address identifying the location of a contact (string) [2-12-15]
  • phone: telephone number for communicating with a contact (string) [2-12-15]
  • email: electronic mailing address for communicating with a contact (string) [2-12-15]
  • location: a place where one or more activities was performed in the development/production/publication (e.g. building/lab) of the chemical analysis (string) [2-26-15]

If you want to prove to the community that the methodology you are reporting is accurate and reliable providing information on method validation is required.  The metadata in this category needs to be open such that the wide range of validation methods can be reported. After thinking about his for a few months the reporting of metrics, statistics, validation processes, or example data (added 02-17-15) is all about making a statement of the quality of the methodology - how good is the method at performing the analysis.  So, the category has been renamed quality metadata (02-17-15). Please let This email address is being protected from spambots. You need JavaScript enabled to view it. know if you have any comments, or additions to the list below.

Metrics

Numeric values that are characteristic of the chemical analysis method

  • Coefficient of Determination (R-squared)
  • Confidence Interval [added 02-17-15]
  • Detection Limit - From Blank Analysis
  • Detection Limit - From Calibration
  • Limit of Quantitation
  • Limit of Linearity
  • Linear Dynamic Range
  • Linear Regression Equation
  • Repeatability (Intra-assay precision)
  • Sensitivity (Slope of the Calibration Curve)
  • Sample Throughput
  • Signal-to-noise ratio (at a specific analyte concentration) [added 02-26-15]
  • Specificity/Selectivity [updated 02-26-15]
  • Stability (remove 02-26-15]

Statistics Chemometrics

Statistical tests that evaluate the appropriateness of the model used in the calibration of data or comparison of chemical analysis metrics between different analysis methods

  • F-test
  • Paired t-test
  • One-way ANOVA
  • Two-way ANOVA
  • Non-parametric Test

Validation

Processes that are carried out to measure accuracy, assess reliability over time or across laboratories

  • Quality Assurance (QA)
  • Quality Control (QC)
  • Statistical Process Control (Control Charts) [moved 02-26-15]
  • SRM/CRM Analysis
  • Sample Spike Recovery
  • Method Blank Analysis (different than the analysis blank)
  • Single Laboratory Validation (Intermediate Precision)
  • Inter-laboratory Validation (Reproducibility) OR Ruggedness

Example Data

  • Chromatogram
  • Peak Table
  • Spectrum
  • Calibration Curve