This shows you the differences between two versions of the page.

Both sides previous revision Previous revision Next revision | Previous revision | ||

ascagenes [2010/03/09 13:15] mjnueda |
ascagenes [2014/05/12 12:37] jcarbonell |
||
---|---|---|---|

Line 7: | Line 7: | ||

Example:\\ | Example:\\ | ||

|Name|Array1|Array2|Array3|Array4|Array5|Array6|Array7|Array8|Array9|…| | |Name|Array1|Array2|Array3|Array4|Array5|Array6|Array7|Array8|Array9|…| | ||

- | |gene1|0.5|0.2|0.7|1.3|1.4|1.0|2.1|2.4|2.6|…. | + | |gene1|0.5|0.2|0.7|1.3|1.4|1.0|2.1|2.4|2.6|…| |

|gene2|0.5|0.3|0.4|0.3|0.4|0.1|0.1|0.4|0.5|…| | |gene2|0.5|0.3|0.4|0.3|0.4|0.1|0.1|0.4|0.5|…| | ||

|...|...|...|...|...|...|...|...|...|...|…| | |...|...|...|...|...|...|...|...|...|...|…| | ||

Line 20: | Line 20: | ||

* //Variability threshold for component selection//: Criterion for selection of principal components. With "average", components are selected that explain more than the average component variability, calculated as total data variability divided by the rank of the matrix associated to the factor. Also a fixed value for percentage of explained variability can be indicated such as 0.2, 0.4, 0.5, 0.7, 0.8...\\ | * //Variability threshold for component selection//: Criterion for selection of principal components. With "average", components are selected that explain more than the average component variability, calculated as total data variability divided by the rank of the matrix associated to the factor. Also a fixed value for percentage of explained variability can be indicated such as 0.2, 0.4, 0.5, 0.7, 0.8...\\ | ||

* //Criterion for gene selection//. Strategy for selection of genes based on SPE and leverage values. Can be either | * //Criterion for gene selection//. Strategy for selection of genes based on SPE and leverage values. Can be either | ||

- | * //Resampling//: where null Leverage distribution are created by permuting columns and genes are selected at the "alpha" percentile of the null distribution and SPE cutoff is computed by using an approximation to a weighted chi-squared distribution [2]. | + | * //Resampling//: where a null Leverage distribution is created by permuting columns of expression data and genes are selected at the "alpha" percentile of the null distribution and SPE cutoff is computed by using an approximation to a weighted chi-squared distribution [[references|[2]]]. |

- | *//alpha//: significant level for gene selection (only for Resampling and Gamma).\\ | + | * //minAS//: where a density function is calculated on the data and genes are selected on a local minimum basis [[references|[7]]]. |

- | *//R//: number of bootstrap rounds for gene selection (only for Resampling)\\ | + | * //Gamma//: where a gamma distribution is adjusted to the distributions of SPE and leverage values, and genes are selected at the "alpha" percentile of the gamma distribution [[references|[7]]]. |

- | * //minAS//: where a density function is calculated on the data and genes are selected on a local minimum basis. | + | |

- | * //Gamma//: where a gamma distribution is adjusted to the distributions of SPE and leverage values, and genes are selected at the "alpha" percentile of the gamma distribution. | + | |

\\ | \\ | ||