Genetic tests  Tests that aim to detect changes in chromosomes, DNA, RNA, protein or metabolites that are associated with e.g. heritable or acquired diseases. They are based on methods from different disciplines such as immunology, molecular biology or cytogenetics.  More  
General Definitions  
Validation  Verification , where the specified requirements are adequate for an intended use. ^{1}  More  
Verification  Confirmation through the provision of objective evidence, that specified requirements have been fulfilled. ^{2}  
Inhouse IVD  Inhouse In Vitro Diagnostic Device; in vitro test which is developed (de novo or modified from a published source) and validated within a medical laboratory where it will be used for the examination of patient specimens.  
Reference Material (RM)  Material, sufficiently homogeneous and stable with respect to one or more specified properties, which has been established to be fit for its intended use in a measurement process. ^{3}  More  
Certified Reference Materials (CRM)  An RM characterized by a metrologically valid procedure for one or more specified properties, accompanied by a certificate that states the value of the specified property, its associated uncertainty, and a statement of metrological traceability. ^{3}  
Inhouse control  A control which has been characterised in the laboratory. It is not a reference material as stability and homogeneity and fitness for intended use have not been assessed.  
Internal control  An internal control is placed in the same reaction tube as the specimen being analyzed. Thus, an internal control will be subjected to exactly the same internal conditions and external parameters as any target sequence(s) present in the tube. A nontarget sequence present in the same sample tube, which is coamplified in order to identify possible inhibition due to thermal cycler malfunction, suboptimal reagents or polymerase activity or the presence of inhibitory substances. Slightly modified from ^{4}  
Traceability  Ability to trace the history, application or location of that which is under consideration. ^{5}  
Traceability chain  Sequence of measurement standards and calibrations that is used to relate a measurement result to a stated metrological reference. ^{1}  
Reference method  For genetic tests, the best available analytical method to detect and/or quantify a particular genotype of interest. For detection purposes, bidirectional sequencing is most often accepted as the reference method.  More  
Gold standard  Nonspecific term indicating that a process or material is the best available approximation of the truth. Use of this term is deprecated. Better use reference standard. ^{4}  
Reference standard  The best available method for establishing the presence or absence of the condition or characteristic of interest. ^{4} ^{6}  
Quantitative method  For molecular genetic tests, an analytical method whose measurement result is quantitative with respect to a target analyte, such as copy number, fragment length, repeat number or concentration; measured indirectly by means of e.g. Ct value, migration distance, peak height, of fluorescent signal etcetera in a sample. Example from Vlassenbroeck2008: The copy number of the methylated MGMT promoter, normalized to the ßactin gene, provides a quantitative test result.  More  
Qualitative method  For molecular genetic tests, an analytical method whose final result is qualitative with respect to a target analyte, such as its presence or absence, in a sample. Note1: Some qualitative tests have no numerical values associated with the results; alternatively, some tests are labeled as qualitative because one of only two results (eg, positive or negative) is reported; these reported results are derived from dichotomizing a quantitative scale. Modified from ^{8}  
Definitions applying to measurements (quantitative)  
Systematic error  An error that is not determined by chance but is associated with the analytical procedure. Predictable. The error occurs in one direction only and is either constant or proportional over the range. See bias .  More  
Random error  An error that results from inconsistent variations from one or more influenced quantities. Associated with the sample, sample matrix, solvents used etcetera. Unpredictable. See precision .  
Uncertainty  Nonnegative parameter characterising the dispersion of the quantity values being attributed to a measurand, based on the information used.  
Accuracy  Closeness of agreement between a measured quantity value and a true quantity value of the measurand. ^{2} Accuracy applies to results and comprises trueness and precision . The reverse of accuracy is uncertainty of measurement, which can be expressed numerically.  Figure 1  
Trueness  Closeness of agreement between the average of an infinite number of replicate measured quantity values and a reference quantity value. ^{2} The reverse of trueness is bias, which can be expressed numerically.  Figure 1  
Bias  Systematic measurement error or its estimate, with respect to a reference quantity value. ^{1}  Figure 1  
Correction factor  The amount of deviation in a measurement that is accounted for in the calibration process. The correction factor can be applied to the measured value, or the measuring instrument can be adjusted.  
Precision  Closeness of agreement between indications or measured quantity values obtained by replicate measurements on the same or similar objects under specified conditions. ^{1} Precision is expressed numerically by measures of imprecision such as a standard deviation or a coefficient of variation.  
Repeatability condition  Condition of a measurement, out of a set of conditions that includes the same measurement procedure, same operators, same measuring system, same operating conditions and same location, and replicate measurements on the same or similar objects over a short period of time. ^{1}  
Reproducibility condition  Condition of a measurement, out of a set of conditions that includes different locations, operators, measuring systems, and replicate measurements on the same or similar objects. ^{1}  
Calibration  Set of operations that establish the quantitative relationship between the instrument, kit or test system measurement signal and the concentration or activity values of an analyte. This relationship is the basis for obtaining a measurement result from the instrument, kit or test system response. It is implicit in the method validation process that the studies to determine method performance parameters are carried out using equipment that is within specification, working correctly, and adequately calibrated.  More  
Calibration curve  
The linear range of analytical instruments, kits, test systems is limited. During method validation the linearity of the calibration curve will be assessed and the (working) range of the calibration curve should be determined.  
Linearity  For the different instruments, kits, test systems used in molecular genetic tests calibration curves will be with respect to different results, such as DNA concentration, fragment length, repeat number …. For example, for a calibration curve with respect to target DNA copy number, linearity means the ability of an analytical method to elicit measurement signals that are directly, or by a well defined mathematical transformation, proportional to the target DNA copy number of analyte in samples within a given range. Mostly a linear regression model is used and the correlation coefficient r used to ascertain goodness of fit of the calibration curve. Mostly a linear regression model is used and the correlation coefficient r used to ascertain goodness of fit of the calibration curve.  More  Method 
Measurement range  Many different related terms and definitions exist (for example range, measurement range, working range, analytical measurement range, linear range, dynamic range, validated range, reportable range). For example for a calibration curve with respect to target DNA copy number, the range will be the interval between the upper and lower copy number values of target analyte in the sample for which the method gives reliable results.Reportable range (CLIA): The range of result values of the analyte for which method performance is reliable and test results can be reported.  Method  
Limit of detection (LOD)  Measured quantity value, obtained by a given measurement procedure, for which the probability of falsely claiming the absence of a component in a material is ß, given a probability a of falsely claiming its presence. ^{2} Notes:
 Method  
Limit of quantification (LOQ)  Method  
Input DNA  With some exceptions as explained below, the input DNA concentration range associated with your test is not a limiting or critical factor nor what you intend to measure (it is a qualitative aspect that interests you, such as the presence of a mutation). Nevertheless too low/high concentrations of input DNA might result in failure to detect anything and each assay will have been validated for a certain input DNA amount or range. Therefore, in many genetic tests DNA concentration is often regarded as a assay condition much like reagent lot, … and variability due to variations in input DNA is investigated during robustness evaluation. Exceptions:
 
Robustness  Measures how resistant testing is to small changes in preanalytic and analytic variables. Such variables may include sample type, sample handling (e.g. transit time or conditions), sample quality, reagent lots or minor changes in assay conditions (e.g. timing or temperature). ^{7} It provides an indication of its reliability during normal use.  Method  
Definitions used in relation with binary test results  
True positive (TP)  A positive result for which the reference method also gives a positive result.  
True negative (TN)  A negative result for which the reference method also gives a negative result.  
False positive (FP)  A positive result while the reference method gives a negative result.  
False negative (FN)  A negative result while the reference method gives a positive result. Remark: At the analytical level TP, TN, FP, FN are used to denote the fractions among the sample population of true/false genotype results. At the clinical level however these annotations denote the fractions among the patient population of true/false test results with regard to disease status. See also definitions of diagnostic sensitivity and diagnostic specificity .  
Sensitivity (analytical) (Se)  The proportion of biological samples that have a positive test result or known mutation and that are correctly classified as positive (assumes mutation is tested for). ^{6} For a specific sample population the calculation for the estimate is : TP/TP+FN Depending on sample size the confidence interval associated with the estimate will vary. More information can be found here. It is regarded good practice to always add a confidence interval with your estimate.  
Specificity (analytical) (Sp)  The proportion of biological samples that have a negative test result or no identified mutation (being tested for) and that are correctly classified as negative. ^{6} For a specific sample population the calculation for the estimate is: TN/TN+FP Depending on sample size the associated confidence interval will vary. More information can be found here. It is regarded good practice to always add a confidence interval with your estimate.  
Confidence interval for proportion  Confidence intervals for proportions are usually set at 95% and based on binomial distributions. xbinci : Excelspreadsheet to calculate exact binomial confidence intervals for sensitivity, specificity or other proportions. TAGS : Estimating sensitivity and specificity without gold standard (maximum likelihood method).  More  
Selectivity  Cross reactivity (profile): The capacity of the method to distinguish the signal of the target sequence from that of other interfering variants/polymorphisms. Interferences: Substances whose presence persists during sample preparation and that influence detection of the target sequence. Interferences (disturbances) might occur in electrophoresis, restriction digestion, blotting, hybridization, sequencing and mass spectrometry based methods of evaluating amplification products.  Method  
Cutoff  The quantitative value of a measured analyte that is used to decide whether the result is considered above or below a clinical or analytical decision point (usually positive or negative). From ^{4} .  
Instrumental cutoff  The response point below which a qualitative test result is determined to be negative and above which the result is determined to be positive (or vice versa). Some qualitative tests have no numerical values associated with the results. Alternatively, some tests are labeled as qualitative because one of only two results (e.g., positive or negative) is reported; these reported results are derived from dichotomizing a quantitative scale [~measurement signal] Through the instrumental cutoff, the results are dichotomized or categorized. For example into mutation present/absent or relative copy number classes..  
Confidence score  Fredd score  
Clinical cutoff  Medical decision point For a test derived from dichotomizing a quantitative or ordinal [~result] scale, there are many possible choices for a cutoff. ^{8} To determine the best medical decision point is a complex in which different have to be taken into consideration. It is often done via ROCanalysis. It follows that for a truly qualitative test, the cutoff is the (only) medical decision point.  
Equivocal zone (=indeterminate zone, gray zone)  It can apply to both analytical and clinical decision making. This zone separates e.g. the ‘target absent’ and ‘target present’ zones. Data that fall within this zone will provide results that are not definitive, such as neither ‘absent’ nor ‘present’ for a specific target; or neither ‘heterozygous’ nor ‘homozygous’ for a specific mutation site. Such results are also called indeterminate results, equivocal results. Modified from ^{4} .  More  
Equivocal result (=indeterminate result, gray zone allele)  It can apply to both analytical and clinical decision making. A test result within a specified range of the cutoff value, which cannot be interpreted as either positive or negative. Note: In molecular genetics, equivocal test results complicate risk interpretations. ^{4} Analytically, indeterminate results are usually retested. Clinically, an equivocal result might be e.g. a ‘gray zone allele’ in repeat expansion disorders for which it is highly unpredictable what the clinical consequences might be.  
Reference range  A particular statistical range, rather than the entire range of observed reference values. Therefore it should be acommpanied by its confidence value (95th percentile range, 99th percentile range, 90th percentile range). Commonly used to characterize the range of test results expected for the nondiseased population. Also called: normal range, reference interval.  More  
Diagnostic sensitivity  The proportion of patients with a specified clinical disorder (or other condition of interest) whose test values are positive or exceed a defined decision limit. Note 1: The clinical disorder must be defined by criteria independent of the test under consideration. Note 2: Is equivalent to clinical sensitivity (US). Note 3: It is the fraction of clinically true positive classifications divided by the sum of clinically true positive and clinically false negative classifications. Note 4: It is the probability (P) that the test is positive (T+) given that the subject being tested is diseased (D+), id est P(T+/D+); or the ability of a test under study to give a positive result for subjects having the disease in question. Modified from ^{4} .  
Diagnostic specificity  The proportion of patients who do not have a specified clinical disorder (or other condition of interest) whose test values are negative or within a defined decision limit. Note 1: The clinical disorder must be defined by criteria independent of the test under consideration. Note 2: Is equivalent to clinical specificity (US). Note 3: It is the fraction of clinically true negative classifications divided by the sum of clinically true negative and clinically false positive classifications. Note 4: It is the probability (P) that the test is positive (T) given that the subject being tested is disease free (D), id est P(T/D); or the ability of a test under study to give a negative result for subjects not having the disease in question. Modified from ^{4} . 




Copyright EuroGentest2 Coordination Action 2011  EU Contract no.: FP7  HEALTHF42010261469