API reference

Every public name exported from pylimma has an entry below; every entry below resolves to a public attribute on the top-level pylimma module. If the two lists ever disagree that is a bug - file an issue.

Linear modelling

pylimma.lm_fit(data[, design, ndups, ...])	Fit linear models to expression data.
pylimma.gls_series(M[, design, ndups, ...])	Fit linear model for each gene using generalized least squares.
pylimma.mrlm(M[, design, ndups, spacing, ...])	Robustly fit linear model for each gene using M-estimation.
pylimma.contrasts_fit(data[, contrasts, ...])	Apply contrast matrix to a fitted model.
pylimma.make_contrasts(*contrasts_args[, ...])	Construct a contrast matrix from contrast expressions.
pylimma.model_matrix(formula, data)	Create a design matrix from a formula and data.
pylimma.e_bayes(data[, proportion, ...])	Empirical Bayes moderation of t-statistics.
pylimma.treat(data[, fc, lfc, trend, ...])	Moderated t-statistics relative to a log fold-change threshold.
pylimma.top_table(data[, coef, number, ...])	Extract a table of top-ranked genes from a linear model fit.
pylimma.top_table_f(fit[, number, genelist, ...])	Top table for multiple coefficients ranked by F-statistic.
pylimma.top_treat(fit[, coef, sort_by, ...])	Top-ranked genes after a treat() fit.
pylimma.decide_tests(data[, method, ...])	Classify genes as differentially expressed.
pylimma.classify_tests_f(fit[, cor_matrix, ...])	Use F-tests to classify vectors of t-statistics into outcomes.
pylimma.squeeze_var(var, df[, covariate, ...])	Empirical Bayes posterior variances.
pylimma.fit_f_dist(x, df1[, covariate])	Fit a scaled F-distribution to sample variances.
pylimma.fit_f_dist_robustly(x, df1[, ...])	Robust estimation of scaled F-distribution parameters.
pylimma.fit_f_dist_unequal_df1(x, df1[, ...])	Fit scaled F-distribution with unequal df1 values.
pylimma.is_fullrank(x)	Check whether a matrix has full column rank.
pylimma.non_estimable(x[, coef_names])	Check for non-estimable coefficients in a design matrix.

Voom and RNA-seq

pylimma.voom(counts[, design, lib_size, ...])	Transform RNA-seq counts for linear modelling with mean-variance weighting.
pylimma.voom_with_quality_weights(counts[, ...])	voom transformation with sample-specific quality weights.
pylimma.vooma(y[, design, block, ...])	voom-like weights for non-count expression data.
pylimma.vooma_lm_fit(y[, design, ...])	Combined vooma + lmFit with iterative refinement.

Normalisation and batch

pylimma.normalize_between_arrays(object[, ...])	Normalize columns of an expression matrix between arrays.
pylimma.normalize_quantiles(A[, ties])	Quantile-normalize columns of a matrix.
pylimma.normalize_median_values(x)	Scale columns so they have the same median.
pylimma.normalize_cyclic_loess(x[, weights, ...])	Cyclic LOESS normalisation of columns of a matrix.
pylimma.background_correct(object[, ...])	Background-correct a single-channel intensity matrix or EList.
pylimma.normexp_fit(x[, method, n_pts, trace])	Estimate parameters of the normal + exponential convolution model.
pylimma.normexp_signal(par, x)	Expected value of signal given foreground under the normal + exponential convolution model.
pylimma.aver_arrays(x[, id, weights])	Average over technical-replicate columns.
pylimma.remove_batch_effect(x[, batch, ...])	Remove batch effects from a matrix of expression values.
pylimma.wsva(y, design[, n_sv, ...])	Weighted surrogate variable analysis.

Duplicates, weights, correlation

pylimma.duplicate_correlation(M[, design, ...])	Estimate correlation between duplicate spots or blocked samples.
pylimma.ave_dups(x[, ndups, spacing, weights])	Average over duplicate spots.
pylimma.avereps(x[, ID])	Average over irregular replicate probes.
pylimma.array_weights(object[, design, ...])	Estimate relative quality weights for each array/sample.
pylimma.array_weights_quick(y, fit, *[, layer])	Compute approximate array quality weights from a linear model fit.

Gene set testing

pylimma.ids2indices(gene_sets, identifiers)	Map named gene sets of identifier strings to zero-based integer indices.
pylimma.roast(y[, index, design, contrast, ...])	Rotation gene-set test (single set).
pylimma.mroast(y, index[, design, contrast, ...])	Rotation gene-set test over many sets.
pylimma.fry(y[, index, design, contrast, ...])	Fast closed-form limit of roast (nrot -> Inf with df.prior=Inf).
pylimma.camera(y, index[, design, contrast, ...])	Competitive gene-set test with inter-gene correlation.
pylimma.camera_pr(statistic, index[, ...])	Pre-ranked competitive gene-set test.
pylimma.inter_gene_correlation(y, design)	Variance-inflation factor and inter-gene correlation.
pylimma.romer(y, index[, design, contrast, ...])	Rotation mean-rank gene-set enrichment analysis.
pylimma.gene_set_test(index, statistics[, ...])	Competitive gene-set test.
pylimma.rank_sum_test_with_correlation(...)	Wilcoxon rank-sum test with an inter-gene correlation adjustment.

GO / KEGG enrichment

pylimma.goana(de, gene_pathway[, universe, ...])	Gene-ontology over-representation analysis.
pylimma.top_go(results[, ontology, sort, ...])	Extract the top GO terms from a goana() result.
pylimma.kegga(de, gene_pathway[, ...])	KEGG pathway over-representation analysis.
pylimma.top_kegg(results[, sort, number, ...])	Extract the top KEGG pathways from a kegga() result.
pylimma.goana_trend(index_de, covariate[, ...])	Estimate per-gene DE probability from a covariate.

Statistical utilities

pylimma.qqt(y[, df, plot_it])	Student's t probability plot (Q-Q plot).
pylimma.zscore_t(x, df[, approx, method])	Z-score equivalents of t-statistics.
pylimma.tricube_moving_average(x[, span, power])	Moving average filter with tricube weights for a time series.
pylimma.convest(p[, niter, plot, report, ...])	Estimate pi0 using a convex decreasing density estimate.
pylimma.prop_true_null(p[, method, nbins])	Estimate the proportion of null p-values.
pylimma.detection_p_values(x, status[, negctrl])	Detection p-values from negative controls.
pylimma.weighted_lowess(x, y[, weights, ...])	Weighted LOWESS smoother - R-compatible entry point.
pylimma.au_roc(truth[, stat])	Area under the empirical ROC curve.

Model selection and mixture models

pylimma.select_model(y, design_list[, ...])	Gene-wise model comparison via AIC, BIC, or Mallows' Cp.
pylimma.fitmixture(log2e, mixprop[, niter, ...])	Fit a mixture model by non-linear least squares.
pylimma.genas(fit[, coef, subset, plot, ...])	Estimate the biological correlation between two contrasts.
pylimma.pred_fcm(fit[, coef, ...])	Predictive (empirical-Bayes shrunken) fold changes.

Splicing

pylimma.diff_splice(fit, geneid[, exonid, ...])	Test for differential exon usage from an exon-level fit.
pylimma.top_splice(fit[, coef, test, ...])	Top-ranked splicing results from a diff_splice fit.
pylimma.plot_splice(fit[, coef, geneid, ...])	Plot exons or isoforms of a chosen gene.

Plotting

pylimma.plot_with_highlights(x, y[, status, ...])	Scatterplot with colour/size highlighting for special groups of points.
pylimma.plot_ma(object[, array, coef, xlab, ...])	MA plot for a matrix, EList, or MArrayLM.
pylimma.plot_md(object[, column, coef, ...])	Mean-difference plot.
pylimma.volcano_plot(fit[, coef, style, ...])	Volcano plot of log-fold-change vs significance.
pylimma.plot_sa(fit[, xlab, ylab, ...])	Sigma vs Amean plot.
pylimma.plot_densities(object[, log, group, ...])	Kernel-density plots of sample intensities.
pylimma.plot_mds(x[, top, labels, pch, cex, ...])	Multidimensional-scaling plot.
pylimma.venn_counts(x[, include])	Cross-tabulate significance indicators.
pylimma.venn_diagram(object[, include, ...])	2- or 3-circle Venn diagram.
pylimma.coolmap(x[, cluster_by, col, ...])	Clustered heatmap with log2-expression colour scheme.
pylimma.barcode_plot(statistics[, index, ...])	Barcode plot of one or two gene sets.

Data classes and dispatchers

pylimma.EList([data])	Expression-list container (Python equivalent of R limma's EList).
pylimma.MArrayLM([data])	Linear-model-fit container (Python equivalent of R limma's MArrayLM).
pylimma.get_eawp(obj, *[, layer, weights_layer])	Polymorphic input dispatcher - port of R limma's getEAWP().
pylimma.put_eawp(slots, original, *[, ...])	Polymorphic output dispatcher - package a result in a form matching the original input's class.