### 1. Highlights

● 2 new directives and 29 new procedures

● Bland-Altman plots to assess the agreement between two variates (`BLANDALTMAN`

).

● procedures for automatic selection of `REML`

random models for single trials, series of trials and meta analysis (`VABLOCKDESIGN`

, `VAMETA`

, `VAOPTIONS`

, `VARANDOM`

, `VAROWCOLUMNDESIGN`

, `VASERIES`

, `VFMODEL`

, `VFSTRUCTURE`

, `VMODEL`

)

● multi-dimensional power models for `REML`

covariance structures, also the linear variance model and the spherical family of geostatistical models (`VSTRUCTURE`

)

● QTL extensions including genomic predictions (`GPREDICTION`

) and analysis of multi-trait trials (`QMTBACKSELECT`

, `QMTESTIMATE`

, `QMTQTLSCAN`

)

● zero-inflated binomial models (`R0INFLATED`

)

● fitting and prediction from Michaelis-Menten curves (`MICHAELISMENTEN`

, `MMPREDICT`

)

● Wald tests in generalized estimating equations (`GEE`

)

● analysis of distance of multivariate data (`MVAOD`

)

● ability to rename data structures (`RENAME`

)

### 2. What’s new

**2.1 Directives**

`QDIALOG`

produces a modal dialog box to obtain a response from the user.

`RENAME`

assigns new identifiers to data structures.

**2.2 Procedures**

`AFAUGMENTED`

forms an augmented design.

`AUSPREADSHEET`

saves results from an analysis of an unbalanced design (by `AUNBALANCED`

) in a spreadsheet.

`BLANDALTMAN`

produces Bland-Altman plots to assess the agreement between two variates.

`DREFERENCELINE`

adds reference lines to a graph.

`EDDUNNETT`

calculates equivalent deviates for Dunnett’s simultaneous confidence interval around a control.

`GPREDICTION`

produces genomic predictions (breeding values) using phenotypic and molecular marker information.

`MICHAELISMENTEN`

fits the Michaelis-Menten equation for substrate concentration versus time data.

`MMPREDICT`

predicts the Michaelis-Menten curve for a particular set of parameter values.

`MVAOD`

does an analysis of distance of multivariate data.

`PLINK`

prints a link to a graphics file into an HTML file.

`QFLAPJACK`

creates a Flapjack project file from genotypic and phenotypic data.

`QMTBACKSELECT`

performs a QTL backward selection for loci in multi-trait trials.

`QMTESTIMATE`

calculates QTL effects in multi-trait trials.

`QMTQTLSCAN`

performs a genome-wide scan for QTL effects (Simple and Composite Interval Mapping) in multi-trait trials.

`QREPORT`

creates an HTML report from QTL linkage or association analysis results.

`QUESTION`

obtains a response using a GenStat menu.

`RFINLAYWILKINSON`

performs Finlay and Wilkinson’s joint regression analysis of genotype-by-environment data.

`SETNAME`

sets the identifier of a data structure to be one specified in a text.

`TCOMBINE`

combines several tables into a single table.

`UTMCONVERSION`

converts between geographical latitude and longitude coordinates and UTM eastings and northings.

`VABLOCKDESIGN`

analyses an incomplete-block design by `REML`

, enabling automatic selection of random and spatial covariance models.

`VAMETA`

performs a `REML`

meta analysis of a series of trials.

`VAOPTIONS`

defines options for the fitting of models by `VARANDOM`

and associated procedures.

`VARANDOM`

finds the best `REML`

random model from a set of models defined by `VFMODEL`

.

`VAROWCOLUMNDESIGN`

analyses a row-and-column design by `REML`

, with automatic selection of the best random and spatial covariance model.

`VASERIES`

analyses a series of trials with incomplete-block or row-and-column designs by `REML`

, automatically selecting the best random models.

`VFMODEL`

forms a model-definition structure for a `REML`

analysis.

`VFSTRUCTURE`

adds a covariance-structure definition to a `REML`

model-definition structure.

`VMODEL`

specifies the model for a `REML`

analysis using a model-definition structure defined by `VFMODEL`

.

**2.2 Functions**

None.

### 3. What’s changed

Most of the changes are compatible with Release 15, the previous release. There are a few commands, however, where new options or parameters have been inserted into the existing lists. These may cause problems in statements where option or parameter names have been omitted or abbreviated (see Section 1.7.1 of Part 1 of the *Guide to the GenStat Command Language* for details). To avoid any difficulty, the name of the option/parameter after the new option/parameter should be given explicitly, and not abbreviated to fewer than four characters.

Any command, where changes in Release 15 may cause incompatibilities in existing programs, is marked in Sections 3.1 and 3.2 by the symbol ^{†}. The full details are given in Section 3.4.

**3.1 Directives**

`SET`

and `GET`

have been extended to allow you to request the use of enhanced computing algorithms from the Intel® Math Kernel Library.

^{†}`VSTRUCTURE`

now allows power models of order one to be defined for points in more than two dimensions, and includes the linear-variance model and the spherical family of geostatistical models.

**3.2 Procedures**

^{†}`AFIELDRESIDUALS`

now lets you reverse the direction of the y-axis, so that the lower coordinates are at the top.

^{†}`AMMI`

has much improved graphics, including AMMI 2 plots.

^{†}`BOXPLOT`

now lets you control the ordering of the boxes when they are plotted horizontally. Also the `ORIENTATION`

option has new settings `horizontal`

and `vertical`

, as synonyms of `across`

and `down`

, to provide better compatibility with other graphics commands, such as `DHISTOGRAM`

and `BARCHART`

.

^{†}`DBIPLOT`

and `GGEBIPLOT`

now allow you to supply your own title.

^{†}`DMSCATTER`

now lets you supply an overall title.

^{†}`DQMKSCORES`

can now produce plots for backcross inbred lines, and the ordering of the `COLOURS`

and `TITLE`

options has been changed to be consistent with the ordering elsewhere (`COLOURS`

before `TITLE`

).

^{†}`DQMQTLSCAN`

and ^{†}`DQSQTLSCAN`

can now plot dominance effects and additive 2 effects.

^{†}`DOTHISTOGRAM`

can now plot the individual dots in different colours, and lets you control the order of the histograms when they are plotted horizontally.

^{†}`GEE`

can now print Wald tests, and has a `FACTORIAL`

option to control the complexity of the model to be fitted.

^{†}`GESTABILITY`

has several more stability coefficients, and can also produce plots of the coefficients against the genotype means.

^{†}`HGFTEST`

now provides more control over the refitting of the models.

^{†}`LVARMODEL`

now allows the structure of the experiment to be specified more conveniently, by supplying block and plot-within-block factors.

`META`

can now save the statistic *Q* for the test of heterogeneity across trials and its number of degrees of freedom, also the random effect variance.

^{†}`MOVINGAVERAGE`

now provides the Holt-Winters method.

^{†}`RCOMPARISONS`

and ^{†}`RTCOMPARISONS`

now provides more control over the way in which the predictions are formed for generalized linear models.

^{†}`RNEGBINOMIAL`

can now save the standard error of the aggregation parameter.

^{†}`RSEARCH`

lets you define a critical value for the probabilities of the terms in a model for that model to be printed.

^{†}`R0INFLATED`

can now fit zero-inflated binomial models, and allows within-group regressions to be done, to save time and space with very large models.

^{†}`QEIGENANALYSIS`

now lets you specify the number of principal components to retain, and to save the number of effective number of columns of the marker data matrix.

^{†}`QMBACKSELECT`

, ^{†}`QMESTIMATE`

, ^{†}`QMQTLSCAN`

, ^{†}`QSBACKSELECT`

, ^{†}`QSESTIMATE`

and ^{†}`QSQTLSCAN`

can now analyse backcross inbred lines.

^{†}`QSASSOCIATION`

has five new options and four new parameters, with enhancements including several methods to define a threshold for significance, the ability to form a kinship matrix, and more efficient analysis of very large data sets.

`R0KEEP`

can now save the Akaike and Schwarz (Bayesian) information coefficients.

`TABSORT`

now lets you take a subset of the table, selecting only the best levels of the factors that are being sorted.

^{†}`TTEST`

now lets you make a permutation test using differences between means rather than t-statistics.

`VTCOMPARISONS`

now provides tests of comparisons, combined over groups.

**3.3 Functions**

None.

**3.4 Incompatibilities**

`AFIELDRESIDUALS` procedure |
option `YORIENTATION` inserted before `PENCONTOUR` . |
---|---|

`AMMI` procedure |
new option `PLOT` replaces the `graph` setting of `PRINT` , and the `GRAPHICS` option is deleted (only high-resolution graphics is now available). |

`BOXPLOT` procedure |
new option `YORIENTATION` inserted after `ORIENTATION` , and options reordered (for compatibility) to move `SCREEN` to come after `METHOD` . |

`DBIPLOT` procedure |
option `TITLE` inserted before `WINDOW` . |

`DMSCATTER` procedure |
new parameter `TITLE` added, and parameters reordered (for compatibility) to become `Y` , `X` , `TITLE` , `YTITLES` , `XTITLES` , `YMARKS` , `XMARKS` . |

`DQMKSCORES` procedure |
ordering of the `COLOURS` and `TITLE` options has been changed, to put `COLOURS` before `TITLE` . |

`DQMQTLSCAN` procedure |
option `POPULATIONTYPE` inserted before `METHOD` and `THRESHOLD` (whose order has been reversed), option `COLOURS` moved to come just before `TITLE` , and parameter `IDEFFECTS` inserted before `IDPARENTS` . |

`DQSQTLSCAN` procedure |
option `POPULATIONTYPE` inserted before `METHOD` and `THRESHOLD` (whose order has been reversed), option `COLOURS` moved to come just before `TITLE` , and parameter `IDEFFECTS` inserted before `IDPARENTS` . |

`DOTHISTOGRAM` procedure |
new option `YORIENTATION` inserted after `ORIENTATION` , and options reordered (for compatibility) to move `SCREEN` to come after `YORIENTATION` . |

`GEE` procedure |
option `FACTORIAL` inserted before `AGGREGATION` . |

`GESTABILITY` procedure |
options `PLOT` , `NBEST` and `DIRECTION` inserted before `PERCENTQUANTILES` . |

`HGFTEST` procedure |
option `LMETHOD` inserted before `DMETHOD` , option `EMETHOD` , inserted before `MLAPLACEORDER` , option `DLAPLACEORDER` inserted before `MAXCYCLE` , and options `EXIT` , `TOLERANCE` , and `ETOLERANCE` inserted before `SAVE` . |

`LVARMODEL` procedure |
parameter `NROWS` deleted, and new parameters `BLOCKS` and `UNITS` inserted before `EFFECTS` . |

`MOVINGAVERAGE` procedure |
option `PRINT` inserted before `NSAMPLES` . |

`QMAP` procedure |
option `TITLE` moved to come just before `OUTFILENAME` . |

`QEIGENANALYSIS` procedure |
option `NROOTS` inserted before `PLOT` , parameter `NEFFECTIVE` inserted before `%VARIANCE` , and default of `SCALING` option changed to `none` |

`QMBACKSELECT` procedure |
option `POPULATIONTYPE` inserted before `ALPHALEVEL` . |

`QMESTIMATE` procedure |
options `NGENERATIONS` , `NBACKCROSSES` and `NSELFINGS` , inserted before `VCMODEL` . |

`QMKDIAGNOSTICS` procedure |
options `NBACKCROSSES` and `NSELFINGS` , inserted before `DCHROMOSOMES` . |

`QMQTLSCAN` procedure |
option `POPULATIONTYPE` inserted before `ALPHALEVEL` , option `COLOURS` inserted before `TITLE` , option `YLABEL` deleted, parameters `IDEFFECTS` and `IDPARENTS` inserted before `QSTATISTICS` (which is the new name for `STATISTICS` ). |

`QRECOMBINATIONS` procedure |
option `TITLE` moved to the end. |

`QSASSOCIATION` procedure |
the options are now `PRINT` , `PLOT` , `RELATIONSHIPMODEL` , , `SCORES` , `METHOD` , `ALPHALEVEL` , `THRMETHOD` , `THRESHOLD` , `DISTANCE` `MINORALLELE` , `KMATRIX` , , `KMETHOD` `SUBPOPULATIONS` , `MODELPART` , `SCALING` , `STANDARDIZE` , , `COLOURS` `TITLE` , `YTITLE` , `XTITLE` , and the parameters are `TRAIT` , `GENOTYPES` , `MKSCORES` , `CHROMOSOMES` , `POSITIONS` , `MKNAMES` , , `IDMGENOTYPES` , `GENFILENAME` , `MAPFILENAME` `WALDSTATISTICS` , `NDF` , `MINLOG10P` , , `LAMBDA` `QSAVE` , `DFILENAME` (with the new or reordered options and parameters shown in italics). |

`QSBACKSELECT` procedure |
option `POPULATIONTYPE` inserted before `ALPHALEVEL` . |

`QSESTIMATE` procedure |
options `NGENERATIONS` , `NBACKCROSSES` and `NSELFINGS` , inserted before `FIXED` . |

`QSIMULATE` procedure |
options `ADD2EFFECTS` and `DOMINANCEEFFECTS` inserted before `QTLCHROMOSOMES` . |

`QMQTLSCAN` procedure |
option `POPULATIONTYPE` inserted before `ALPHALEVEL` , option `COLOURS` inserted before `TITLE` , parameters `IDEFFECTS` and `IDPARENTS` inserted before `QSTATISTICS` (which is the new name for `STATISTICS` ). |

`QTHRESHOLD` procedure |
option `POPULATIONTYPE` inserted before `THRMETHOD` , parameters `ADD2PREDICTORS` and `DOMINANCEPREDICTORS` inserted before `THRESHOLD` . |

`QUESTION` |
The `QUESTION` directive has been replaced by the `QDIALOG` directive, with updated facilities appropriate to the more recent computing environments. The new `QUESTION` procedure uses `QDIALOG` to duplicate most of the facilities of the directive, so that existing programs can still run. The main difference is that the `SAVE` option and the option settings `MODE=e` and `MODE=f` are no longer supported. |

`RCOMPARISONS` procedure |
options `WEIGHTS` , `OFFSET` , `METHOD` , `ALIASING` , `BACKTRANSFORM` , `SCOPE` , `NOMESSAGE` , `DISPERSION` , `DMETHOD` and `NBINOMIAL` inserted before `LSDLEVEL` . |

`RNEGBINOMIAL` procedure |
option `SEAGGREGATION` inserted before `MAXCYCLE` . |

`RSEARCH` procedure |
option `PPROBABILITY` inserted before `FINALMODELS` . |

`RTCOMPARISONS` procedure |
options `WEIGHTS` , `OFFSET` , `METHOD` , `ALIASING` , `BACKTRANSFORM` , `SCOPE` , `NOMESSAGE` , `DISPERSION` , `DMETHOD` , and `NBINOMIAL` inserted before `LSDLEVEL` . |

`R0INFLATED` procedure |
options `XGROUPS` and `ZGROUPS` inserted before `MAXCYCLE` ; parameter `NBINOMIAL` inserted before `RESIDUALS` . |

`TTEST` procedure |
option `PLOT` inserted before `NTIMES` , and option `PERMMETHOD` inserted before `SEED` . |

`VSTRUCTURE` directive |
parameters `DISTANCES` and `COORDINATES` inserted before `INITIAL` . |