**Absorbance.** Absorbance, or optical density, is a measure of the amount of light absorbed by a solution. Absorbance is equal to the logarithm of the ratio of incident light to transmitted light.
**Accuracy.** Accuracy is the nearness of a measurement to its true value or accepted value.
**Adjusted Variance.** The adjusted variance is the variance of the mean of the adjusted responses from replicate samples.
**Adjusted Response.** The adjusted response is the raw response of a sample, adjusted to remove signal variation. It is a measurement of the total ligand binding of the sample. The adjusted response is obtained by multiplying the raw response times the multiplying factor for that assay run. The multiplying factor is calculated by dividing the relative activity entered in Test Method Setup by the average of the tracer activity raw responses. By adjusting the tracer activity (sometimes termed total count in isotopic assays) to the relative activity, signal variation (differences from assay run to assay run due to decay, enzyme/substrate deterioration, instrument drift, etc.) can be factored out from binding variation, (differences from assay run to assay run due to binder deterioration, incubation differences, buffer changes, etc.). An arbitrary value (relative activity) is entered to which the average tracer activity's response will be adjusted. This value is defaulted to 100,000 CPM for isotopic and chemiluminescent assays, and to 2 for colorimetric test methods. If tracer activity samples are assayed, StatLIA will adjust all raw responses to this value to remove signal variation. These adjusted responses will then be used for all computations.. If tracer activity samples are not assayed, no adjustment will be performed and the raw responses will be used for all computations.
Top of Page
**Adjusted CV.** The adjusted CV is the coefficient of variation (standard deviation/mean) of the adjusted responses from replicate samples.
**Affinity.** Affinity is a measure of the intrinsic binding strength of the ligand binding reaction. The intrinsic attractiveness of the binder for the ligand is typically expressed as the equilibrium constant (Ka) of the reaction. The equilibrium constant Ka = [Ligand-Binder]/[Ligand][Binder], where [ ] represents the molar concentration of the material at equilibrium.
**Amplification.** Amplification is the use of substances which directly increase the amount of signal measured in proportion to the quantity of analyte.
**Analysis of Variance**. Analysis of variance (ANOVA) is a statistical test which compares the distribution of two or more sample groups to determine if one or more of the groups are significantly different from the others. The sample groups must be gaussian in distribution. In an analysis of variation, the average variance within each of the sample groups is factored out from the variance between each of the sample groups before computing the probability of significant differences between the groups.
**Analyst.** The analyst is the individual who performs the assay run.
**Analyte.** An analyte is the compound being measured. In immunoassays, the analyte can be either the ligand or the binder.
**Antibody**. An antibody binder is an immunological protein which binds to a specific antigen (ligand).
**Antigen.** An antigen is a ligand that contains a region or epitope which is specifically recognized by an antibody binding site.
**Antiserum.** Antiserum is blood serum which contains antibodies.
Top of Page
**Archival.** Archive data files contain the data and some of the protocol specifications from a set of assay runs. The archive data files are stored separately from the test data files. Archived assay runs cannot be recomputed or included in a quality assurance evaluation unless they are retrieved back into the test data files.
**Assay Table.** The assay table is a table listing pertinent information about the newly defined assay runs awaiting computation. Once a new assay run has been successfully computed and saved in the test data file, the assay run is deleted from the assay table. Some of the information in this table can be modified and the sequence of assay runs can also be changed.
**Assay Run.** An assay run is a set of standard samples, control samples, and unknown samples which are analyzed and computed together in a single batch.
**Assay Date.** The assay date is the date the assay run was performed. The assay date is automatically entered in LIA Setup. In isotopic tests the assay date, along with the expiration date, are used to calculate what the activity of the tracer activity sample would be on the expiration date.
**Assay Information Table.** An assay information table is a set of reference or current assays which contains fifteen items of information specific for each assay run.
**Assay Worklist Report.** An assay worklist report is a listing of the placement of the baseline standard samples, standard samples, control samples, and unknown samples. The assay worklist report also includes unknown record information and other information relating to a specific assay run.
**Assay Set Table.** The assay set table is a listing of all thirty reference assay runs and all thirty currrent assay runs. The assay runs are identified by assay name and assay date.
**Assay Set.** An assay set is a group of assay runs comprising all of the reference assays (maximum 30 assay runs) or all of the current assays (maximum 30 assay runs).
Top of Page
**Assay Name.** The assay name is the first three letters of the test name followed by a hyphen and the assay number. The assay number is the next sequential assay number and is assigned automatically. All data pertaining to that assay run is referenced by the assay name. The assay name cannot be changed.
**Assay Data File.** Assay data files are temporary files which contain all of the data necessary to compute a new assay run. All assay runs listed in the Assay Table (accessed from the main screen) or the assay box in Assay Run Setup (Modify) have corresponding assay data files. When a new assay run has been successfully computed and saved in the test data file, the assay data file is deleted.
**Assay Box.** An assay box is a selection box listing all of the assay runs and nonimmunoassay batches currently defined and awaiting computation or printing. It is used to select an assay run to review, modify, or delete. Once a new assay run has been successfully computed and saved in the test data file, the assay run is deleted from the assay box.
**ASTM.**
**Asymmetrical Standard Curve.** An asymmetrical standard curve is a sigmoidal curve with one asymptote more elongated than the other asymptote when the response of each standard is plotted against the logarithm of its respective concentration.
**Asymptote.** The asymptotes of the sigmoidal immunoassay standard curve are the two ends of the curve which plateau on the response axis as the curve approaches infinitely high and infinitely low standard concentrations.
**Available.** Individual workstations can be made available or unavailable for the computation of new assay runs. When all of the raw data has been collected from the detector instrument for an individual assay run, the assay data is sent for computation to the first workstation set at Available which is not performing any tasks. The order in which workstations are queried for availability is highest workstation number to lowest number (Station 01, the primary workstation).
Top of Page
**Avidity.** Avidity is a measure of the association intensity of the reactions of the binder. Avidity is influenced by all of the intrinsic ligand-binder affinities such as the multiple interactions found with polymeric antibodies and multivalent antigens binding at multiple contact points.
**Axis.** An axis is a vertical (ordinate) or horizontal (abscissa) dimension which intersects the other dimension at the origin of a plot.
**Background.** Background samples are instrument blanks which measure the amount of the raw response which is due to residual activity from the detector. This residual activity can be from instrument noise, extraneous radiation, and other artifacts. The average of the background sample raw responses is subtracted from all sample raw responses prior to all subsequent calculations. The raw response which is stored in test data includes the background average response.
**Baseline Standards.** The baseline standards are samples which are used in the mathematical calibration of all samples but are not plotted on the standard curve. Baseline standards include background samples, tracer activity samples, nonspecific binding samples, and maximum binding samples. The last two baseline standards are sometimes used to represent the infinitely low and infinitely high standard concentrations at the two asymptotes.
**Batch Name.** The batch name is the identifier which is transmitted from a detector instrument and is used to identify all of the raw data from a specific assay run.
**Beer-Lambert Law.** The Beer-Lambert Law is an equation which describes the linear relationship between the absorption and the concentration of the absorbing material (absorber). The law states that absorbance (ABS) is equal to absorptivity a, multiplied by the pathlength or the distance the light travels through the sample b, multiplied by the concentration c (ABS= abc).
Top of Page
**Beta Particles.** Beta particles are subatomic particles, essentially electrons, which are emitted during the disintegration of certain radioactive isotopes such as 3H and 14C, which are used as tracers in radioimmunoassays. Beta emitting samples are solubilized in scintillation cocktails containing fluorescent organic compounds and measured in liquid scintillation counters. Many compounds, including water, reduce the fluorescent efficiency by absorbing some of the radiation. This quenching activity can be corrected using quench correction curves or other correction mechanisms. Many liquid scintillation counters automatically correct for quench.
**Bin.** Bins are regions of the standard curve bordered midway between the mean responses of adjacent standards. The response is either the adjusted response or the normalized response. Each standard curve is divided into the same number of bins as there are standards. If tracer activity samples are assayed, screening tests are divided into ten bins with the borders at ten percent intervals of the mean response of the tracer activity. Binning allows the response or concentration of standard and unknown samples to be examined individually at specific regions of the standard curve. Binning is also used to compare the variance lines of the responses from standards with the variance line of the responses from unknowns.
**Binder.** A binder is a substance which is capable of binding specifically and reversibly with a ligand. The principle types of binders are antibodies (immunoassays), naturally occurring binding proteins (competitive protein binding assays), and naturally occurring cell receptors (receptor assays).
**Binding Type.** The binding type is the type of ligand immunoassay technology used in the test method, such as RIA or ELISA.
**Bmax.** The baseline standard expected to yield the highest possible bound response is termed the Bmax. In competitive binding assays, the Bmax is the Bzero baseline standard. In sandwich assays, the Bmax is the maximum binding baseline standard. If a Bmax baseline standard is not assayed, the raw responses can be normalized to any standard. In one standard assays, the Bmax is the index standard.
Top of Page
**Bmin.** The baseline standard expected to yield the lowest possible bound response is termed the Bmin. In competitive binding assays, the Bmin is the NSB, a zero standard with no binder added. In sandwich assays, the Bmin is the zero blank baseline standard. If a Bmin baseline standard is not assayed, the raw responses can be normalized to any standard. In one standard assays, the Bmin is the reagent blank.
**Bound Fraction.** The bound fraction of an immunoassay reaction is portion containing all of the ligand-binder complexes.
**Buffer.** The buffer is the defined media in which the ligand-binder reaction takes place. Many of the components in the buffer solution influence the avidity of the reaction.
**Check Point.** Security check points are specified procedures within the program which can have access restricted to authorized analysts only and can have all access to that procedure recorded.
**Chemiluminescence.** Chemiluminescence is the production of light photons by a chemical or electrochemical reaction. Chemiluminescence usually involves the oxidation of an organic compound such as luminol or acridinium esters by an oxidant such as hydrogen peroxide or hypochlorite. Chemiluminescent reactions occur in the presence of catalysts such as alkaline phosphatase, horseradish peroxidase, metal ions or metal complexes.
**Chi Square Test.** The Chi-Square test is a statistical test which computes the probability that there is no significant difference between the expected frequency of an occurrence with the observed frequency of that occurrence.
**Chi Square Distribution.** Chi square, or goodness of fit, distributions are a family of probability distributions of the Chi-Square statistic, one for each degree of freedom. For small degrees of freedom, the distribution is skewed to the right. As the number of degrees of freedom increases, the distribution rapidly becomes symmetrical. For large degrees of freedom, the chi square distribution closely approximates a gaussian curve.
Top of Page
**Coefficient of Variation.** Coefficient of variation, or CV, is a statistical measurement of the distribution of responses around the mean of those values. The CV is only applicable for values which come from a gaussian distributed population. It is derived by dividing the standard deviation of the replicate adjusted or normalized response measurements by the mean of the replicate measurements (times 100 for percent). Since the measurement units (e.g. CPM) are canceled out, the %CV is a relative value that is independent of the magnitude of the replicate response measurements.
**Colorimetric Label.** A colorimetric label is an enzyme system in which an enzyme linked to a ligand or binder reacts with a specific substrate to generate a chromophore product. This tracer conjugate is quantified by measuring the amount of light absorbed by the product at a specific wavelength. Enzymes commonly used to label ligands or binders include alkaline phosphatase using para-nitrophenyl phosphate as substrate, horseradish peroxidase using hydrogen peroxide/coupler as substrate, and b-galactosidase using o-nitrophenylgalactoside as substrate.
**Comparison Probability.** The comparison probability (Cpr Prob) is the probability that the assay run's standard curve is not significantly different from the reference assay standard curves. In the case of a logistic curve fit, the comparison probability should be evaluated in conjunction with the parallel probability. Linear regressions are straight lines and a comparison probability is not computed.
**Competitive Binding.** A competitive binding assay is an immunoassay reaction which is based upon the competition of labeled and unlabeled ligand for a limited number of binding sites on the binder. A fixed amount of labeled ligand (tracer) and a variable amount of unlabeled ligand are incubated with the binder. Following the law of mass action, the amount of labeled ligand which can bind to the binder is a function of the total concentration of labeled and unlabeled ligand. As the concentration of unlabeled ligand is increased, less labeled ligand can bind to the binder and the measured response is decreased. The standard curve of a competitive binding assay has a negative slope and is generally symmetrical at the midpoint.
**Concentration Scale.** The analyte concentration scale is the scale used to graph the analyte concentration on the x-axis. The analyte concentration can be graphed on a logarithmic scale or on a linear scale.
**Concentration Units.** The concentration units are the units of measurements used to quantify the analyte.
Top of Page
**Concentration.** The concentration of an analyte is a measurement of the mass of the analyte per unit volumn. The mass may be expressed in grams, moles, international units (IU), or other units. The liquid volumn may be expressed in milliliters or other units of volumn. Sometimes only the total mass of the analyte is used. One standard test results are often the ratio of sample to index standard response alone. The index standard can have a predetermined concentration which is then multiplied times the ratio for a concentration in whatever units are entered.
**Confidence Range.** The confidence range is a numerical interval for a single parameter. This numerical interval is the range which encompasses a specified percentage of the corresponding parameter from the reference assay runs.
**Conjugate.** See Tracer.
**Control Specimen.** A control specimen is material from a single pool of unknown specimen(s) from which an aliquot is measured every assay run to monitor assay to assay reproducibility and to provide an indirect measurement of the performance of the assay components.
**Control Group.** A control group is one of three sets of parameters which measure the performance of a test method. The three sets of control parameters are the method controls, the control specimens, and the subpopulation distributions.
**Counts Per Minute.** Counts per minute (CPM) are the number of radioactive disintegrations which were recorded by a detector instrument during one minute.
**Crossreactivity.** Crossreactivity is a measurement of the reactivity between a binder and a different ligand than the ligand to which the binder is specific.
**Current Assays.** The set of current assay runs consists of the last thirty assays performed and provides a running group of parameters that can be compared to the reference set to describe the recent performance of the assay run.
Top of Page
**Curve Fit.** Curve fit algorithms are mathematical formulas which compute a continuous curve function using discrete coordinate points.
**Cutoff.** Subpopulation cutoff values are concentration boundaries which separate a subpopulation into negative (less than the cutoff value) and positive (equal to or greater than the cutoff value) groups.
**Decay Constant.** The decay constant is a constant unique to each radioactive nuclide which represents the proportion of the atoms in a sample of that radionuclide undergoing decay in unit time. The decay constant of a nuclide is related to the nuclide's halflife.
**Degrees of Freedom.** The degrees of freedom of a statistical group are the number of values in the group which are free to vary. This number is usually one less than the sample size, the number of items in the group.
**Denaturation.** Denaturation is the loss of the native configuration of the macromolecule, such as the unfolding of the tertiary structure of an antibody protein. Denaturation usually results in the loss of the macromolecule's biological or immunological reactivity or solubility.
**Detector Data.** The detector data are the initial sample measurement(s) recorded by the detector instruments which is converted into the raw data. The raw data are then transmitted from the detector instrument.
**Detector. **A detector is an analytical instrument which is capable of measuring the amount of tracer label in an immunoassay sample, such as scintillation counters, microplate readers, and automated immunoassay analyzers.
**Dialog Box.** A dialog box is an on-screen window which appears when certain functions are selected. The dialogue box requests additional information or selections.
**Dilution Factor.** See Dilution.
Top of Page
**Dilution.** A dilution is the ratio of the volume of pure specimen to the total volume of specimen plus a diluent such as buffer. Dilutions are expressed as the number of parts of pure specimen, a colon, and the total number of parts in the solution. The number of parts of diluent are determined by subtracting the number of pure specimen parts from the total. A dilution of 1:10 contains one part pure specimen and nine parts diluent. A dilution factor is the total number of parts with the volume of pure specimen being one part.
**Disintegrations Per Minute.** Disintegrations per minute (DPM) are the number of radioactive disintegrations which actually occurred during one minute. The DPM of a sample is usually determined by dividing the number of disintegrations which were recorded by the efficiency of the detector instrument under the same conditions, particularly the amount of quenching, as were present in the measured sample.
**Dose.** See Concentration.
**Emission Wavelength.** See Excitation Wavelength.
**Emission Spectrum.** The emission spectrum of a substance such as a fluorophore is the intensity of the electromagnetic radiation from the substance as a function of the excitation wavelength of the absorbed light.
**Empirical Based Curve Fitting.** An empirical based standard curve is a method of data reduction which treats all data points as discrete elements and constrains the curve to pass through the mean of each standard response. Empirical methods make the assumption that each standard response mean is the true response of that standard.. Each standard point has few degrees of freedom and a low level of statistical reliablility. Empirical methods cannot compensate for the decreased reproducibility at different regions of the standard curve. Empirical methods do not compensate at all for outlier error. Empirically derived standard curves can not be compared statistically to the reference assay standard curves.
Top of Page
**End Point Measurements.** End point measurements are measurements of an enzyme-substrate reaction, which are made at a fixed time, usually after the reaction has been completed.
**End Point Immunoassay.** In end point immunoassays, the signal is measured when the ligand-binder reaction has reached effective equilibrium. Any nonspecific component of the signal is assumed to be small and constant after the reaction has reached equilibrium.
**Enzyme Immunoassay.** An enzyme immunoassay (EIA) is a heterogeneous competitive binding immunoassay in which the tracer material is an enzyme label.
**Enzyme-Linked Immunosorbant Assay.** An enzyme linked immunosorbant assay (ELISA) is a heterogenous enzyme immunoassay in which one of the reaction components (ligand or binder) is adsorbed nonspecifically to a solid phase surface. The analyte to be measured is added, followed by an enzyme labeled binder conjugate. This forms a sandwich complex of binder-analyte-conjugate or ligand-analyte-conjugate attached to the solid phase which can be measured when the enzyme substrate is added to the bound conjugate. The standard curve of an ELISA assay has a positive slope and is often asymmetrical.
**Enzyme Activity.** Enzyme activity is a measure of the rate, or velocity, of the conversion by the enzyme of the substrate to the product per unit time.
**Enzyme System.** An enzyme system is a solution containing all of the necessary substrates and cofactors for an enzyme reaction. Sometimes two or more enzymes are used which act in concert or in sequence on a substrate to convert the substrate to a certain product. The product can be a chromophore, a fluorophore, or other molecule having physical properties which can be detected and quantified.
Top of Page
**Enzyme Multiplied Immunoassay Technique.** Enzyme multiplied immunoassay technique (EMIT) assays are homogenous competitive binding enzyme immunoassays in which a low molecular weight ligand is attached to an enzyme. Enzyme activity is inhibited when the enzyme labeled conjugate is bound to a binder. Addition of unlabeled ligand analyte results in a competition for binding sites on the binder and increasing amounts of free enzyme conjugate as the concentration of unlabeled ligand increases. The standard curve of an EMIT assay has a positive slope and is often asymmetrical.
**Enzyme.** An enzyme is a protein which catalyzes a specific chemical reaction.
**Equilibrium Constant.** The equilibrium constant (Km) is the intrinsic affinity constant of the reaction between a binder and a ligand.
**Equilibrium.** Equilibrium is a condition of a ligand-binder reaction in which the rate of the formation of the ligand-binder complex is equal to the rate of disassociation of the ligand-binder complex back to free ligand and free binder.
**Error Message.** An error message is a warning which is displayed whenever an error occurs during the operation of the program. A record of the type of error, its location, and its cause is stored in the Error Log.
**Estimated Dose.** Estimated doses are monitor points that give the calculated concentration for predetermined responses on the y-axis. An associated number, such as ED50, refers to the percentage of the maximum response (Bmax) for which a concentration is computed.
**Evaluation Weight.** An evaluation weight is a numerical multiplier which allows each control group to have a greater or smaller contribution to the final quality control probability index, relative to the other control groups.
Top of Page
**Excitation Wavelength.** Excitation wavelength in fluorometry is the wavelength of radiant energy that is absorbed by a fluorophore molecule. This incident light excites the molecule and results in the emission of radiation at a higher wavelength (the emission wavelength).
**Excluded.** Excluded unknowns are unknown specimens which are not included in any subpopulation distribution analysis. Excluded samples are identified by an Ex- before the specimen's sequence number.
**Execution Error Log.** The execution error log is a record of all errors encountered during the operation of the program. Information is recorded listing the type of error, its location, and its cause.
**Expected Precision.** The predicted variance is the response variance that is expected at a single response point. This response variance is computed from a variance line derived from the means and variances of the reference assay responses. The heteroscedasticity of the test method, the random error, and the systematic error are the principle elements factored into the variance line.
**Expiration Date.** The expiration date of the reagent kit (or the tracer material) is established by the manufacturer as the last date that the reagents will not have degraded appreciably. For quality control purposes, the expiration date of isotopic immunoassays is used with the assay date to calculate what the activity of the tracer activity samples would be on the expiration date.
**Export Assay Data.** Export Assay Data is a utility used to electronically transfer computational and quality control data from individual assay runs to other databases.
Top of Page
**F Distribution.** The F distribution is a family of distributions, one for each pair of degrees of freedom from an F test ratio. The distributions are generally asymmetrical and skewed to the right when the two degrees of freedom are low. The distributions become more symmetrical as the two degrees of freedom increase.
**Fit Probability.** The fit probability (Fit Prob) is the probability that the computed standard points on the logistic curve are not significantly different from the observed points. The fit probability is derived from the residual variance.
**F Test.** An F test is a statistical test which computes the probability that the ratio of two variances from two different populations are equal.
**Flag.** A flagged parameter is a parameter that is outside of a specified probability limit. A flagged parameter indicates a possible problem with one or more components in the test method.
**Fluorescence Polarization Immunoassay.** Fluorescence polarization immunoassays (FPIA) are homogenous competitive binding immunoassays in which fluorophore-labeled ligand competes with the unlabeled ligand for binding sites on the specific binder. When excited with polarized light, fluorophore ligands bound to binder rotate more slowly and emit light at a higher intensity than do free labeled ligands.
**Fluorescence.** Fluorescence is the property of certain molecules, or fluorophores, to absorb light at one wavelength and emit a light at a longer wavelength. The incident light excites the molecule to a higher level of vibrational energy. As the molecule returns to the ground state, the excited fluorophore emits a photon. This photon is the fluorescence emission. If the molecule returns to the ground state through an intermediate excited triple state, there is a delay in the emission of the photon. This delayed photon is termed a phosphorescence emission.
Top of Page
**Fluorescent Label.** A fluorescent label is either a fluorescent molecule, or an enzyme system that generates a fluorescent product, which is linked to a ligand or binder tracer. The tracer is quantified by measuring the amount of light emitted by the tracer at a specific wavelength after exciting the tracer with incident light of a shorter wavelength. Fluorescent molecules include fluorescein and lanthanide chelates such as europium, samarium, and terbium.
**Fluoroimmunoassay.** Fluorescence immunoassays (FIA) are competitive binding immunoassays in which the analyte content of the sample is measured by the amount of fluorescence from bound labeled ligand. FIA's can also use a fluorogenic enzyme to tag the tracer and measure the formation of a fluorescent product.
**Fluorophore.** See Fluorescence.
**Formatted Raw Data.** The formatted raw data are the sample response data obtained directly from the detector instrument and processed into a standardized format.
**Free Fraction.** The free fraction of an immunoassay reaction is the fraction of ligand which is not bound to the binder. When the free fraction is measured in isotopic tests, the average raw response of the (free) samples is subtracted from the raw response average of the tracer activity samples to yield the bound raw response. The bound raw response is then used to calculate the adjusted, normalized and y-axis responses.
**Frequency Distribution.** A frequency distribution is a systematic way to order a set of data from lowest to highest value showing the number of occurrences (frequency) at each value or range of values.
Top of Page
**Gamma Radiation.** Gamma radiation is a type of high energy electromagnetic radiation which is emitted during the disintegration of certain radioactive isotopes such as 125I and 57Co, which are used as tracers in immunoassays. Gamma emitting samples are measured in solid crystal scintillation counters. These solid crystals, are usually thallium activated sodium iodide and are different sizes to accommodate lower and higher energy radioisotopes.
**Gaussian Distribution.** A gaussian, or normal, distribution is a symmetrical frequency distribution having a precise mathematical formula relating the mean and standard deviation of the samples. Gaussian distributions yield bell shaped frequency curves having a preponderance of values around the mean with progressively fewer observations as the curve extends outward.
**Halflife.** The halflife of a specific radioactive isotope is the length of time necessary for half of the radioactive nuclei in a sample to disintegrate. The halflife of a radioisotope is unique for each isotope.
**Heterogenous Immunoassay.** Heterogenous immunoassays are reactions in which a physical separation step is required to separate the bound ligand from the free ligand before either fraction can be measured.
**Heteroscedastic Variances.** Heteroscedastic variances are unequal variances. One of the fundamental characteristics of most immunoassays is that the variances of the sample response change throughout the range of the standard curve.
**Highest Specific Binding.** The highest specific binding (HSB) standard is the standard expected to yield the highest bound response. The HSB is the internal response which is used to calculate the normalized response if the appropriate baseline standard was not assayed. The HSB is also used to calculate one of the method control monitor points in the absence of the appropriate baseline standard. In competitive binding assays, the lowest concentration standard is the HSB. In sandwich binding assays, the highest concentration standard is the HSB.
Top of Page
**Homogenous Immunoassay.** Homogeneous immunoassays are assays which use differences between bound and unbound label signals to measure the bound or the free fractions without the need to separate the two fractions.
**Identification Fields**. Three identification fields are used to identify the unknown specimens and are common to all test methods. The first field is the sequence number of each unknown record. In unknown files, the sequence number is preceded by Uk- for all unknown specimen concentrations which are included in subpopulation distributional tests. Unknown specimen concentrations which are not included in subpopulation distributional tests are designated as excluded samples and are identified by an Ex- before the specimen's sequence number. The second field contains the concentration results for each unknown specimen. The third field contains the identification of the unknown specimen. The identification field is used to merge to LIM files when dual labeled assays are measured.
**Immunoassay.** An immunoassay is a ligand binding assay which uses an antibody and the antibody's corresponding antigen as the binder and ligand. The term immunoassay is often expanded to include any procedure in which the quantitation of an analyte depends on the progressive saturation of a specific binder or ligand, either of which may be the analyte, and the subsequent determination of the analyte's distribution between the bound and free fractions. This includes competitive protein binding assays, where the binder is a naturally occurring binding protein, and receptor assays, where the binder is a naturally occurring cell receptor.
**Immunofluorometric Assay.** Immunofluorometric assays (IFMA) are sandwich assays in which one binder (or ligand) is used to capture the analyte and then a second binder or ligand conjugate, tagged with a fluorescent compound or fluorescent-converting enzyme, binds to the analyte. This reaction forms a sandwich complex of binder-analyte-conjugate or ligand-analyte-conjugate which is then measured. The standard curve of an IFMA assay has a positive slope and is often asymmetrical.
Top of Page
**Immunometric Assay.** Immunometric assays are immunoassays in which the tracer is a labeled binder. Most immunometric assays involve two binders, both present in excess. The first binder is usually absorbed onto a solid phase and is used to capture the ligand (analyte). The second binder, the tracer, binds to another epitope on the ligand analyte. Both binders are present in large excess. These assays are sometimes called sandwich binding assays. Presumably, all of the analyte is bound in these binder-ligand-binder complexes. As the concentration of analyte is increased, the measured response is increased. The standard curve of an immunometric assay has a positive slope and is often asymmetrical.
**Immunoradiometric Assay.** Immunoradiometric assays (IRMA) are immunometric assays in which one binder (or ligand) is used to capture the analyte and a second binder, tagged with a radioisotope, then binds to the analyte. This forms a sandwich complex of binder-analyte-conjugate or ligand-analyte-conjugate which is then measured. The standard curve of an IRMA assay has a positive slope and is often asymmetrical.
**Incubation.** The incubation of an immunoassay is the time during which the ligand binding occurs.
**Index Standard.** The index standard is a baseline standard which is used to calculate the sample's response ratio in one standard tests. The index standard's adjusted response can either be the numerator or the denominator to the sample's adjusted response. This ratio is then multiplied by the concentration of the index standard for the sample concentration.
**Interferant.** An interferant is a substance which alters the ligand-binder reaction or the label signal.
Top of Page
**In Use.** The (In Use) message that appears next to a test name in the test method selection box occurs under two conditions. Either the test data file is being used by another workstation or the file was not closed properly. To reset the test, verify that it's not being used by another workstation and select Instrument/Reset. Select the test method and click on Select at the bottom of the selection box.
**Isotope.** An isotope is a nuclide with the same atomic number but different atomic mass. Unstable radioactive isotopes are used for labeling in radioimmunoassays (RIA) and immunoradiometric assays (IRMA).
**Kinetic Analysis.** A kinetic analysis measures the change in concentration of the starting and end products in a ligand-binder reaction or an enzyme-substrate reaction over time.
**Kruskal-Wallis.** The Kruskal-Wallis test is a nonparametric analysis of variance which ranks the values of two or more sample groups from lowest to highest and then compares the distribution of these ranks to determine if one or more of the groups are significantly different from the others.
**Kurtosis.** Kurtosis is a measurement of the peakedness (broad or narrow) of a frequency distribution.
**Label.** The label is an atom or molecule which is attached to either the ligand or binding protein and which, by itself or as part of an enzyme system, is capable of generating a signal which can be quantitated.
Top of Page
**Laboratory Address.** The laboratory address is a three line address will be used to identify the laboratory and the workstation. The serial number of the workstation appears on the first line of the laboratory address on the Laboratory Setup screen and on Laboratory Setup Reports. The laboratory address without the workstation serial number appears on every other report and on the StatLIA Manager screen.
**Leptokurtic Distribution.** A leptokurtic distribution is a gaussian distribution having a positive kurtosis and a narrow shaped peakedness.
**Ligand Excess.** Ligand excess is the presence of excess ligand, in relation to the binder concentration. Severe ligand excess can result in increased solubility of ligand-binder complexes, decreased apparent reactivity, and underestimation of the ligand quantity.
**Ligand.** A ligand is a substance which is capable of binding specifically and reversibly with a binder. A ligand is termed an antigen when the binder is an antibody.
**LIM Name.** The LIM name is the computer file name of a LIM file. The LIM name is used to identify the computer file containing all of the unknown records whose corresponding unknown specimens are assayed together in a single assay run.
**LIM Result Log.** The LIM result log is a record of the actions taken on all LIM files. The LIM result log records the assay run which assayed the unknown specimens along with whether the LIM file was was deleted or was sent to the LIM system, and the date, time and person who sent or deleted the LIM file.
**LIM Text Format.** StatLIA supports three LIM text formats in a LIM file to separate the individual fields in each unknown record. The three formats supported are comma delimited, tab delimited, and set field lengths (space delimited). A LIM text format must be selected prior to receiving and sending LIM files.
Top of Page
**LIM Field.** A LIM field is one of eight identification or classification fields in an unknown record which contains a single item of information pertaining to an individual unknown. The first three LIM fields are permanent fields and are common to all test methods. The last five fields are separately defined in each individual test method.
**LIM File.** A LIM file is a computer file containing all of the unknown records with all of the identification and classification information for all of the unknown specimens assayed together in a single assay run. A LIM file is the file which is transferred to and from the LIM system. An unknown file is created from the corresponding LIM file for use within StatLIA. A LIM file is created from the corresponding unknown file to transfer back to the LIM system.
**Linear Regression Coefficients.** A least squares linear regression is a model based equation which describes a straight line. It is defined with two coefficients. Coefficient a is the y-intercept; the point on the y-axis that the line intersects. Coefficient b is the slope of the line.
**Linear Binding Region.** The linear binding region of an immunoassay standard curve is any region (usually midrange) which appears linear when the standard responses are plotted against their respective concentrations. The length of this region is only a portion of the usable concentration range of the test method and the length of this region (if present) varies among test methods. A linear binding region apparent in an immunoassay standard is not due to any fundamental relationships like the spectral properties of substances in the Beer-Lambert Law.
Top of Page
**Logistic.** A five parameter logistic curve is a model based equation which describes a nonlinear curve such as that found with most immunoassay standard curves. Logistic curves are defined with five coefficients. Coefficient a is the high asymptote. Coefficient b directs the shape of the curve. Coefficient c is the midpoint of the curve. Coefficient d is the low asymptote. Coefficient g is the asymmetry parameter. When the curve is symmetrical and the point of inflection is at the midpoint of the curve, coefficient g equals 1 and the equation becomes a four parameter logistic.
**Logit.** The logit transform is a transformation which stretches the interval between successive responses of sample concentration logarithms disproportionally at the two asymptotic ends of the standard curve and makes the standard curve more linear. A logit transform also disproportionally increases the heteroscedasticity at the two asymptotes.
**Lot Number.** The lot number is information which identifies the reagents used in an assay run. If there is no master lot number for all of the reagents, the lot number and the expiration date of a single reagent are entered. For isotopic assays, the radiolabeled reagent is the reagent recorded.
**Lowest Specific Binding.** The lowest specific binding (LSB) standard is the standard expected to yield the lowest bound response. The LSB is used to calculate one of the method control monitor points in the absence of the appropriate baseline standard. In competitive binding assays, the highest concentration standard is the LSB. In sandwich binding assays, the lowest concentration standard is the LSB.
**Mann-Whitney Test.** The Mann-Whitney test is a nonparametric statistical test which computes the probability that the rank values of two sets of a single parameter are members of the same population. The ranking of the original parameter data is lowest to highest. The Mann-Whitney test is the nonparametric equivalent of the t test and is used when the data are not sufficiently gaussian in distribution or when the variances of the two groups are too unequal.
Top of Page
**Manufacturer.** The manufacturer is the company or source of the reagents used for all of the assay runs in a test method. If the reagents in a test method are from many different companies, a single reagent vendor may be entered.
**Matrix.** The matrix of an immunoassay reaction is everything in the incubation medium except the ligand and the binder. Many of these matrix components influence the kinetics of the ligand-binder reaction.
**Mean.** Mean is the arithmetic average of a set of data and is a measurement of the central tendency of that data.
**Mesokurtic Distribution.** A mesokurtic distribution is a gaussian distribution having no kurtosis, or peakedness.
**Method Controls (MC).** The method controls are standard curve parameters that measure, directly or indirectly, the actual analytical methodology itself. The method (standard curve) controls consist of the adjusted and normalized standard responses, standard curve coefficients, monitor points (min/max detectable concentrations, ED 20/50/80, etc.), dilution curve measurements, and replicate precision probabilities. From the method control parameters, assessments can be made about the performance of the label material, the tracer, the binder, the standards, the buffer matrix, the incubation conditions, and the separation of bound and free ligand. The individual method control parameters are listed in the statistical quality control tables from an assay report. All method control probabilities are weighted equally in the probability index calculation.
Top of Page
**Model Based Curve Fitting.** A model based standard curve is a method of data reduction which combines all standard determinations to derive one mathematical equation to describe every point on the curve. Model based methods do not make the assumption that each standard response mean is the true response of that standard. By combining all standard determinations, model based methods have more degrees of freedom and therefore a high level of statistical reliability. Model based methods allow each standard point to be weighted according to the reliability of the point at that region of the curve. Model based methods reduce the distortion caused by outliers. Model based standard curves can be compared statistically to the reference assay standard curves.
**Monitor Point.** A monitor point is a mathematical determination of a physical property of an assay run which can be used as a gauge to measure the performance of some aspect of that assay run. Baseline monitor points such as %NSB/TA (the amount of nonspecific binding as a percentage of the total tracer activity) and %MB/TA (the maximum amount of binding as a percentage of the total tracer activity) provide fundamental information about two absolute binding conditions. Estimated doses such as ED20, ED50, and ED80 measure the location of the standard curve at specific percentages (20%, 50%, 80%, respectively) of the maximum binding obtained that assay run. By using a value normalized to an internal point, binding condition variations which affect the whole assay run are factored out of the standard curve measurements and individual performance at specified points on the standard curve can be monitored.
**Monoclonal Antibody.** Monoclonal antibodies are immunoglobulins which arise from a single clone of B-lymphocyte cells. Monoclonal antibodies are usually obtained by fusing a single B-lymphocyte with a hybridoma tumor cell.
**Mouse.** A mouse is a computer input device that controls the movement of the cursor on the screen to make selections and to perform operations.
**Multiplying Factor.** The multiplying factor is a number which, when multiplied times the raw response, yields the adjusted response. The multiplying factor is used to factor out the signal variation from assay to assay from the binding measurements recorded for each sample. The multiplying factor is determined by dividing the average of the tracer activity's raw response by the relative activity.
Top of Page
**Multipoint Analysis.** Multipoint analysis is the measurement of a specimen at various dilutions of the specimen.
**Negative.** When a cutoff concentration value has been entered for a subpopulation reference range, specimens which yield concentrations below that value are reported in the unknown sample table of an assay report as negative.
**Network.** A network is a set of hardware and software that allows more than one computer to share the same programs, data files, and peripheral equipment.
**Nonimmunoassay.** A nonimmunoassay test (NIA) is a single set of samples which are measured on a detector instrument as a single batch but is not computed as an immunoassay. The measurements are printed as formatted raw data in a single report.
**Nonisotopic Assays.** Nonisotopic assays are immunoassays which do not have radioactive labels for their tracer material.
**Nonparametric Tests.** Nonparametric tests are statistical tests which make no assumptions about the distribution of a statistical population. Nonparametric tests are generally more robust but less sensitive than parametric tests.
**Nonspecific Binding.** Nonspecific binding is the measurement of label or labeled conjugate present in the bound fraction of a ligand-binder reaction which occurred from nonimmune-specific means. Nonspecific binding is the only type of binding found in competitive binding immunoassay samples (NSB) having no binder present; and in sandwich binding immunoassay samples (Zero Blank) having no ligand present. Nonspecific binding samples as baseline standards can be used to represent infinitely high standard concentrations (competitive binding) or infinitely low standard concentrations (sandwich binding).
Top of Page
**Normal Distribution.** See gaussian distribution.
**Normalized Response.** The normalized response is the percentage of bound ligand-binder complex of the sample relative to a specified baseline standard, or standard, response within the assay run. Whereas the adjusted response measures the actual amount of ligand-binder binding which occurred in the sample, the normalized response measures the amount of binding relative to another sample within the assay run. This means that all of the binding variation between assay runs, which is due to binder deterioration, incubation differences, buffer changes, etc., are included in the adjusted response. These binding differences between assay runs are factored out of the normalized response. This allows the integrity of the sample itself to be examined without being distorted by differences in the binding conditions.
**Normalized Variance.** The normalized variance is the variance of the mean of the replicate samples' untransformed normalized response.
**Normalized CV.** The normalized CV is the coefficient of variation (standard deviation/mean) of the replicate samples' untransformed normalized response.
**Nuclide.** A nuclide is the nucleus of an atom having a specific atomic number and atomic mass number. A nuclide may be radioactive.
**One Standard Test.** A one standard test is an immunoassay test in which only one standard is assayed. This baseline standard, called the index standard, is used to calculate the ratio of each sample to the index standard response. The index standard's adjusted response can be either the numerator or the denominator to the sample's adjusted response. The ratio is then multiplied by the concentration entered for the index standard.
Top of Page
**Parallel Probability.** The parallel probability (Par Prob) is the probability that the shape, or form, of the assay run's standard curve is not significantly different from the shape of the reference assay standard curves. A parallel probability is only computed for logistic curves. Linear regressions are straight lines and a parallel probability is not computed.
**Parallelism.** Parallelism is the extent to which two or more different substances produce parallel dilution (titration) curves in an immunoassay. Parallelism measurements are used to detect matrix effects in different substance solutions. Parallelism is also used to measure the crossreactivity of a binder to substances similar in structure to the ligand.
**Parameter.** A parameter is a quantifiable characteristic or feature of a statistical population.
**Patient.** See Unknown.
**Photon.** A photon is a particle consisting of a discrete packet of radiant energy.
**Platykurtic Distribution.** A platykurtic distribution is a gaussian distribution having a negative kurtosis and a broad shaped peakedness.
**Poisson Distribution.** A poisson distribution is a distribution of random occurrences in which one occurrence has no influence on any other occurrence. The variance of a poisson distribution is equal to its mean and therefore the standard deviation is equal to the square root of the mean of the distribution. Radioactive decay measurements follow a poisson distribution and therefore have a lower measurement error when more counts are accumulated.
Top of Page
**Polarized Light.** Polarized light is light that travels in a single plane.
**Polyclonal Antibody.** Polyclonal antibodies are a pool of immunoglobulins which arise from more than one clone of B-lymphocyte cells. Polyclonal antibodies are usually present when antisera from a conventional immunization is used.
**Population.** A statistical population, or universe, is the entire collection of measurements of the variable being studied.
**Positive.** When a cutoff concentration value has been entered for a subpopulation reference range, specimen concentrations equal to or greater than the cutoff value are reported in the unknown sample table of an assay report as positive.
**Possible Outlier.** A possible outlier is a parameter probability which is equal to or lower than the possible probability limit but higher than the probable probability limit. Possible outlier parameters have shifted enough from the reference assay group to warrant inspection but are not as severely shifted as a probable outlier.
**Precision Probability.** The precision probability is the computed probability resulting from a statistical comparison between the reproducibility of the specimen replicate responses (measured as variance or CV) and the reproducibility predicted for that response point. The predicted reproducibility is calculated from a variance line computed from the variances or CV's of the reference assays in Test Method Analysis. A separate variance line is computed for the standards and for the unknowns. Expected control sample variances are computed from the unknown variance line.
**Precision PI.** A precision PI is a probability index computed using the individual precision probabilities of the sample responses from each baseline standard and standard specimen, control specimen, and unknown specimen.
Top of Page
**Precision.** Precision is a measurement of the reproducibility of replicate sample responses in an assay run. Precision is composed of both the random error and the systematic error of the measurement.
**Primary Workstation.** The primary workstation in a StatLIA System is the workstation which is directly connected to the BayTech multiplex buffer or to a detector instrument. The primary workstation is the only workstation which can collect raw data from the BayTech or a detector instrument. The primary workstation processes this raw data and then directs the completed assay data files to the available workstations in the system for computation. The primary workstation is identified as Station 01.
**Probability Limits.** A probability limit is a probability which forms a boundary between the acceptance (greater than) or rejection (less than) of the hypothesis that there is no significant difference between the groups. The possible probability limit signals a parameter which may be compromised. The probable probability limit signals a parameter which is severely shifted.
**Probability Index.** The probability index (PI) for a control group is calculated by averaging the inverse of all the probabilities within the group and then inverting this average.
Top of Page
**Probability.** Probability is the likelihood that the difference between two or more groups of the same parameter is due to random error alone. Probabilities are expressed as a decimal value between 0 and 1, which is the calculated probability of no significant difference between the parameters. A high probability means that there is no significant difference between the parameters. A low probability signals a significant difference between the parameters. The probabilities listed in StatLIA's reports are the probabilities of no significant difference between a paramter in one or more assay runs and the corresponding parameter from the reference assays. Probability limits are used to define the probability values, below which the parameter being tested is considered significantly different from the reference assays.
**Probable Outlier.** A probable outlier is a parameter probability which is equal to or lower than the probable probability limit. Probable outlier parameters are severely shifted and require inspection.
**Protocol Specifications.** Protocol specifications are the physical details which define the test method. Protocol specifications include sample information such as the number of standard and their respective concentrations, the number of replicates assayed for each unknown specimen, and the placement of control samples. Protocol specifications also include information necessary for the data reduction and quality control and analysis of assay runs, such as the typed of curve fitting selected, the variance line coefficients, and subpopulation categories.
**Quality Assurance.** Quality assurance is a comparison between the control parameters from a group of current assays and the corresponding parameters from the reference assays.
**Quality Control.** Quality control is a comparison between the control parameters from one assay run and the corresponding parameters from the reference assays.
**Radioactivity.** Radioactivity is the emission of subatomic particles and radiation from elemental isotopes having unstable atomic nuclei. The isotopes used in radioimmunoassays emit either beta particles or gamma radiation. This radioactive decay is a random event and follows a poisson distribution.
**Radioimmunoassay.** Radioimmunoassay (RIA) is a heterogeneous competitive binding immunoassay in which the tracer material is a radioactive isotope.
Top of Page
**Random Error.** Random error is the component of the total error which is due to chance alone.
**Ratio.** A ratio is a mathematical formula which divides a numerator by a denominator. Normalized responses, baseline monitor points, and screening test results are all examples of ratios.
**Raw Response.** The raw response of a sample is the raw data of that sample, in the form which it is used for the computation of the assay run. The average of the instrument background raw response is subtracted from raw response of each sample before any computations are performed. The raw response, minus the background raw response, is stored in the appropriate test data and archive files.
**Raw Data.** The raw data are the unprocessed sample response data which are transmitted directly from the detector instrument.
**Reagent Blank.** A reagent blank is a baseline standard sample from a one standard screening test which measures the amount of nonspecific activity present in the reaction. The amount of nonspecific activity is measured by omitting the binder in labeled ligand (competitive binding) assays or the ligand in labeled binder (sandwich binding) assays.
**Recovery.** The recovery of an analyte is an analytical procedure which is used to measure the inaccuracy of the test method. Recovery is measured by assaying a sample after adding a known amount of the analyte and the same sample without added analyte.
Top of Page
**Reference Assays.** The reference assays are a fixed set of stable assays which are statistically compared to a single assay or multiple assays to detect changes in reagents or incubation conditions. These reference assays, by default, are the first 30 assays run on StatLIA for each test method. With a reference set of at least two assays, standard curve and control specimen parameters in today's assay are statistically compared to the same parameter in the reference assays to identify any statistically significant differences.
**Reference Range. **The reference range of a subpopulation is an analyte concentration range encompassing a defined percentage of all of the members of that subpopulation.
**Reference Curve.** The reference curve is a single standard curve computed from the standard responses of all of the reference assays.
**Reference Cutoff.** A reference cutoff is an analyte concentration which divides the unknowns within a subpopulation into negative (those specimen concentrations less than the cutoff value) and positive (those specimen concentrations equal to or greater than the cutoff value).
**Relative Activity.** Relative activity is an arbitrary response value to which the raw response of the tracer activity sample average is adjusted to remove signal variation. The relative activity is divided by the average raw response of the tracer activity baseline standard to calculate a multiplying factor for each assay run. All raw responses in an assay run are multiplied by the assay's multiplying factor to yield the adjusted response. The variation in the raw response of the tracer activity between assay runs is due to signal variation and is not a component of the variation in the ligand-binder reaction. Adjusting the raw responses to the adjusted responses factors out the signal variation and only the binding variation remains.
**Replicate.** See Sample.
Top of Page
**Report Logs.** Report logs are files in which summary information from several different program operations are recorded.
**Response Expression.** The response expression is one of five forms of the sample response which is used to calculate the y-axis of standard curve and to compute all sample concentrations (x-axis). The five response expressions are the adjusted response, the normal response, the log (base e) of the adjusted response, the log (base e) of the normalized response, and the logit response.
**Response.** Response is a generic term which refers to any form of the tracer label measurement recorded by the detector instrument.
**Results.** A result is the computed concentration of a specific analyte in a specimen. A specimen concentration less than or greater than the minimum or maximum concentration entered in Test Method Setup is recorded as < or > the minimum or maximum concentration, respectively. A specimen concentration inside of the minimum and maximum concentration range but less than or greater than the lowest or highest standard concentration, respectively, is recorded with a ? followed by the computed concentration.
**Sample Blank.** An unknown sample blank is an aliquot from a specimen which is used to measure the amount of nonspecific activity or the amount of endogenous analyte present in the specimen. The response from the activity blank is subtracted from the response of the sample before computing the specimen concentration. The concentration of the concentration blank is subtracted from the concentration of the sample to generate the specimen concentration.
Top of Page
**Sample.** A sample, or replicate, is an aliquot of a baseline standard specimen, standard specimen, control specimen, or unknown specimen, and all of the reagents which comprise an individual immunoassay incubation medium.
**Sandwich Assay. **See Immunometric Assay.
**Scatchard Analysis.** A Scatchard analysis is a computation of the number of binding sites and the affinity constants for each subclass of binder present plus a measurement of the amount of nonspecific binding. When the proper Scatchard model is chosen, the coefficients from a Scatchard analysis of a single assay provide direct measurements of the number of binding sites and the affinity constant of each binder subclass, which are then statistically compared to those from the reference assay set.
**Secondary Workstation.** A secondary workstation in a StatLIA System is any workstation which is not connected to the BayTech multiplex buffer or a detector instrument. The primary workstation is the only workstation which can collect raw data from the BayTech or a detector instrument. Secondary workstation are identified by station numbers greater than Station 01.
**Security Log. **The security log is a record of all accesses to active security check points. The security log records the check point procedure, analyst, date, time, and other information specific for each check point.
**Selected Button.** The selected button is a button on the screen which changes color or shading to indicate that that function has been chosen.
**Sensitivity.** The sensitivity of an test method is the lowest concentration that can be reliably measured using that test method.
**Separation.** Separation is the process of physically separating the bound ligand from the free ligand before either fraction can be measured.
Top of Page
**Sequence Number.** The sequence number of an unknown record is the order each unknown is placed in an unknown file or a LIM file. The sequence number is the first permanent field in an unknown file or a LIM file.
**Serial Number.** A serial number is assigned to each copy of the StatLIA program for identification and registration purposes. The serial number of the workstation appears on the first line of the laboratory address on the Laboratory Setup screen and on Laboratory Setup Reports.
**Signal Variation.** Signal variation is the variation in the measurement of the tracer label from assay run to assay run. Signal variation is due to many factors including detector instrument drift, isotopic decay, enzyme denaturation, and substrate degradation. Signal variation will distort the binding variation from assay run to assay run if it is not factored out.
**Signal.** The signal is the physical activity of the labeled tracer material which is measured by a detector instrument. The signal is the response which is measured for each sample.
**Skewness.** Skewness is an asymmetrical frequency distribution in which the values are concentrated on one side of the central tendency and trail out on the other side. If the trail is to the right or positive end of the scale, the distribution is said to be positively skewed. If the distribution trails off to the left or negative side of the scale, it is said to be negatively skewed.
**Slope.** The slope of a line is the tangent of that line. The tangent is the vertical distance of any two points on a line divided by the horizontal distance of the same two points. The slope in a linear regression is expressed as coefficient b.
Top of Page
**Solid Phase.** Binder or ligand can be absorbed onto solid phase surfaces, such as microtiter wells, magnetic particles, or plastic beads. This immobilized reactant can then capture ligand or binder from the liquid phase. The soluble free fraction can then be washed away.
**Specific Activity.** The specific activity of a tracer is the activity of the label per unit mass of the labeled material.
**Specificity.** The specificity of a binder is the ability of its binding site to distinguish between the ligand to which the binder is specific and other compounds.
**Specimen.** A specimen is standard calibrator material, known control material, or material from an unknown subject which is collected for subsequent measurement in an immunoassay. An aliquot of specimen, together with the other components of the incubation medium, are added to a sample for assay.
**Spectrophotometer.** A spectrophotometer is a detector instrument that measures the amount of monchromatic light passing through a solution by means of an adjustable monochromator such as a prism or diffraction grating.
**Standard Curve.** An immunoassay standard curve is a curve (or straight line) produced by mathematically fitting an equation to the data from a series of dilutions of an analyte of known concentration. The data is a plot of the response versus the concentration of each dilution.
**Standard Deviation.** The standard deviation is the square root of the variance.
Top of Page
**Standard.** Standard samples are aliquots of calibration specimens containing predetermined quantities of the analyte. The response of each standard, along with the standard's predetermined concentration, is used to construct a standard curve. From this standard curve, sample concentrations can be computed using the response from the sample.
**Subpopulation Distributions.** A subpopulation distribution is a frequency distribution of the analyte concentrations of all of the members of a subset of unknowns grouped according to related characteristics (e.g., sex and age).
**Substrate.** See Enzyme System.
**Symmetrical Standard Curve.** A symmetrical standard curve is one which yields a sigmoidal curve with both asymptotes elongated equally when the response of each standard is plotted against the logarithm of its respective concentration. The inflection point of a symmetrical standard curve, where the mirror image of one half of the curve is superimposable on the other half, is the midpoint of the curve.
**Systematic Error.** Systematic error is the component of total error which is due to changes in the test method.
**T Distribution.** The t distribution (sometimes called Student's t distribution) is a statistical distribution which is used when a gaussian distribution is not applicable. The t distribution is a series of distributions which approach a gaussian distribution the larger the sample size but are more leptokurtic the smaller the sample size. The standard deviation of a t distribution with less than an infinite sample size is less than that of the gaussian distribution.
Top of Page
**T Test.** The t test is a statistical test which computes the probability (p) that two groups of a single parameter are members of the same population. The population must follow a t distribution. However, a t test is robust enough to allow considerable deviation from normality.
**Test Data File.** The test data files contain all of the protocol information and data for a single test method including all of the data from the thirty reference assay runs and the thirty current assay runs. The archive data files are stored separately from the test data files.
**Test Method.** A test method is an immunoassay procedure developed to measure a specific analyte. The term test method encompasses the protocol specifications such as the type of curve fit selected and the number of unknown replicates, the reagents used in the assay runs, and all of the data generated from all of the assay runs of that test method. There can be more than one test method for a single analyte.
**Test Code.** The test code is a code consisting of numbers or letters which can be used as an alternate identification for a test method.
**Test Name.** The test name is the name used to identify a single test method. The test name must be between three and eight characters long. The first three characters in a test name must be unique within a single StatLIA System. These three letters form the first three characters of an assay name (followed by a hyphen and the next sequential number). Everything about a particular test method is referenced by the test name.
**Test Defined LIM Fields.** The five test defined LIM fields are used to classify the unknown subjects for subsequent grouping with the appropriate subpopulation from a test method. The test defined LIM fields may be different from test method to test method. Test defined LIM fields may also be used for additional identification or to transmit footnotes back to the LIM system.
Top of Page
**Time-Delayed Fluorescence.** Time-delayed fluorescence is a technique in which the fluorescence of slowly emitting compounds, such as metal chelates, is measured.
**Titer.** The titer of a binder is the dilution factor used to bind a specific ligand concentration range.
**Total Count.** See Tracer Activity.
**Total Error.** Total error is composed of random error, which is the error attributable to chance, and systematic error, which is the error attributable to changes in the test method.
**Tracer Activity. **The tracer activity samples are baseline standards which measure the activity of the tracer label without a ligand-binder factor. The tracer, substrate, and buffer measured in the tracer activity samples are taken from the same sources that are used for the rest of the same assay run. The tracer activity samples from isotopic tests are sometimes termed total count samples. Isotopic tracer activity samples contain the same amount of tracer that is added to each assay sample and are therefore higher in activity than any other sample. Nonisotopic tracer activity samples might not have the highest activity of all samples if, for example, the enzyme/substrate mixture must be diluted to be measured. The bound raw response is then used to calculate the adjusted, normalized and y-axis responses. The variation in the tracer activity raw response from assay to assay is a measure of the label response variation. Label response variation is caused by anything which affects the raw response, such as instrument drift, enzyme/substrate deterioration, and isotopic decay, that is not related to changing conditions in the ligand-binder reaction.
**Tracer.** The tracer is a ligand or binder which has been attached to a measurable label.
**Transformed Data.** Transformed data are data which has been converted mathematically in order to better fit certain statistical or mathematical models.
Top of Page
**Unavailable.** See Available.
**Unknown File.** An unknown file is a StatLIA file of all of the unknown records containing all of the identification and classification information of all of the unknown specimens assayed together in a single assay run. The first field of each unknown record contains the sequence number preceded by either Uk- or Ex-. A LIM file is generated from the corresponding unknown file which was transmitted from the LIM system.
**Unknown Record.** An unknown record is a set of eight LIM fields, each of which contains a single item of classification information pertaining to the same unknown. The first three LIM fields are permanent fields and are common to all test methods. The last five fields are separately defined in each individual test method.
**Unknown.** An unknown is the subject from which a specimen is collected for subsequent measurement in an immunoassay. An unknown is identified and classified in a LIM file or unknown file.
**Unselected Button.** The unselected button is a button on the screen which is colored or shaded to indicate that that function has not been chosen.
**Variance Model.** The variance model is one of four least squares linear regression types. The four types of linear regression are the linear, exponential, logarithmic, and the power line fits.
**Variance Expression.** The variance expression of the response, computed as a function of the mean response of a variance line, is either the variance or the coefficient of variation (CV) of the responses at a single point.
Top of Page
**Variance Line.** A variance line is an equation which describes the average variance of a response point as a function of the mean of that response. The response variance is computed from the means and variances of the reference assay responses. The heteroscedasticity of the test method, the random error, and the systematic error are the principle elements contributing to the variance line.
**Variance.** The variance of a parameter pool is a measure of the variability of the pool. It is determined by squaring the deviations of each value about the mean and dividing by one less than the number of individuals in the pool. The variance is the square of the standard deviation.
**Wavelength.** A wavelength is the linear distance which is traversed by one complete wave cycle of electromagnetic energy.
**Weighting.** Weighting is a mathematical process which allows individual values in a data set to be given greater or less impact in an equation than their values alone would contribute.
**Zero Standard.** The zero standard is a baseline standard calibrator which contains no detectable concentration of the analyte being assayed. In competitive binding assays, the zero baseline standard is termed the Bzero. In sandwich assays, the zero standard is termed the zero blank.
**Zero Blank.** See Zero Standard
Top of Page |