topicks.hpp File Reference

Choose top genes for downstream analyses. More...

#include <vector>
#include <algorithm>
#include <numeric>
#include <cstddef>
#include <type_traits>
#include "sanisizer/sanisizer.hpp"

Include dependency graph for topicks.hpp:

Go to the source code of this file.

Classes
struct	topicks::PickTopGenesOptions< Stat_ >
	Options for pick_top_genes(). More...

Functions
template<typename Stat_ , typename Top_ , typename Bool_ >
void	topicks::pick_top_genes (const std::size_t n, const Stat_ *const statistic, const Top_ top, const bool larger, Bool_ *const output, const PickTopGenesOptions< Stat_ > &options)
template<typename Bool_ , typename Stat_ , typename Top_ >
std::vector< Bool_ >	topicks::pick_top_genes (const std::size_t n, const Stat_ *const statistic, const Top_ top, const bool larger, const PickTopGenesOptions< Stat_ > &options)
template<typename Index_ , typename Stat_ , typename Top_ >
std::vector< Index_ >	topicks::pick_top_genes_index (const Index_ n, const Stat_ *const statistic, const Top_ top, const bool larger, const PickTopGenesOptions< Stat_ > &options)

Detailed Description

Choose top genes for downstream analyses.

Function Documentation

pick_top_genes() [1/2]

template<typename Stat_ , typename Top_ , typename Bool_ >

void topicks::pick_top_genes	(	const std::size_t	n,
		const Stat_ *const	statistic,
		const Top_	top,
		const bool	larger,
		Bool_ *const	output,
		const PickTopGenesOptions< Stat_ > &	options )

Template Parameters

Stat_	Numeric type of the statistic for picking top genes.
Top_	Integer type of the number of top genes.
Bool_	Output boolean type.

Parameters

	n	Number of genes.
[in]	statistic	Pointer to an array of length n, containing the statistics with which to rank genes.
	top	Number of top genes to choose. Note that the actual number of chosen genes may be: smaller than top, if the latter is greater than the total number of genes in the dataset. smaller than top, if PickTopGenesOptions::bound is set and top is greater than X, where X is the number of genes in the dataset with statistics greater than PickTopGenesOptions::bound (or less than the bound, if PickTopGenesOptions::larger = false). larger than top, if PickTopGenesOptions::keep_ties = true and there are tied statistics at the top-th chosen gene.
	larger	Whether the top genes are defined as those with larger statistics.
[out]	output	Pointer to an array of length n. On output, the i-th element will be true if gene i is one of the top genes and false otherwise.
	options	Further options.

pick_top_genes() [2/2]

template<typename Bool_ , typename Stat_ , typename Top_ >

std::vector< Bool_ > topicks::pick_top_genes	(	const std::size_t	n,
		const Stat_ *const	statistic,
		const Top_	top,
		const bool	larger,
		const PickTopGenesOptions< Stat_ > &	options )

Template Parameters

Bool_	Output boolean type.
Stat_	Numeric type of the statistic for picking top genes.
Top_	Integer type of the number of top genes.

Parameters

	n	Number of genes.
[in]	statistic	Pointer to an array of length n, containing the statistics with which to rank genes.
	top	Number of top genes to choose, see the pick_top_genes() overload for more details.
	larger	Whether the top genes are defined as those with larger statistics.
	options	Further options.

Returns

A vector of booleans of length n, indicating whether each gene is to be retained.

pick_top_genes_index()

template<typename Index_ , typename Stat_ , typename Top_ >

std::vector< Index_ > topicks::pick_top_genes_index	(	const Index_	n,
		const Stat_ *const	statistic,
		const Top_	top,
		const bool	larger,
		const PickTopGenesOptions< Stat_ > &	options )

Template Parameters

Index_	Integer type of the output indices.
Stat_	Numeric type of the statistic for picking top genes.
Top_	Integer type of the number of top genes.

Parameters

	n	Number of genes.
[in]	statistic	Pointer to an array of length n containing the statistics with which to rank genes.
	top	Number of top genes to choose, see the pick_top_genes() overload for more details.
	larger	Whether the top genes are defined as those with larger statistics.
	options	Further options.

Returns

Vector of sorted and unique indices for the chosen genes. All indices are guaranteed to be non-negative and less than n.