astyping.cpp File Reference

#include <iostream>
#include <fstream>
#include "RNAModel.h"
#include <mysql++.h>
#include "dbinfo.h"
#include <algorithm>

Classes
class	ASCountKeeper
Functions
void	statAStype (Connection &conn, const string &dbname, const string &resfile, bool use_raw=true)
int	astypes (const Noschain &ch1, const Noschain &ch2, int tc[12], map< int, int > &ritype, ASCountKeeper &ack, map< int, int > &ril, map< int, int > &nril, set< pair< int, int > > &majorRI, set< pair< int, int > > &minorRI)
int	skipExon (const Noschain &lch, const Noschain &sch, int &i, int &j, ASCountKeeper &ack)
int	skipExonVariant (const Noschain &ch1, const Noschain &ch2, int &i, int &j, ASCountKeeper &ack)
bool	before (const Noschain &ch1, const Noschain &ch2, const int i, const int j)
int	retainIntron (const Noschain &lch, const Noschain &sch, int &i, int &j, ASCountKeeper &ack, set< pair< int, int > > &retained)
bool	adskip (const Noschain &ch1, const Noschain &ch2, int &i, int &j)
bool	bothADAA (const Noschain &ch1, const Noschain &ch2, int &i, int &j, ASCountKeeper &ack)
bool	alternativeDonor (const Noschain &ch1, const Noschain &ch2, int &i, int &j, ASCountKeeper &ack)
bool	alternativeAcceptor (const Noschain &ch1, const Noschain &ch2, int &i, int &j, ASCountKeeper &ack)
bool	CEorAA (const Noschain &ch1, const Noschain &ch2, int &i, int &j, ASCountKeeper &ack)
bool	mutualExclusive (const Noschain &ch1, const Noschain &ch2, int &i, int &j, ASCountKeeper &ack)
void	usage ()
int	main (int argc, char *argv[])
Variables
const char *	ASTYPE [12] = {"AD", "AA", "CE", "CE_AD", "CE_AA", "CE_ADAA", "IR", "ADAA", "CEorAAend", "ME", "Other", ""}

---

Function Documentation

bool adskip	(	const Noschain &	ch1,
		const Noschain &	ch2,
		int &	i,
		int &	j
	)

this has bee replace with better function

this function has been replace with better version

References Range::begin(), Range::direction(), Range::end(), and Noschain::numberOfRanges().

Referenced by astypes().

bool alternativeAcceptor	(	const Noschain &	ch1,
		const Noschain &	ch2,
		int &	i,
		int &	j,
		ASCountKeeper &	ack
	)

References ASCountKeeper::addAA(), Range::begin(), Range::end(), Range::isPlus(), Noschain::numberOfRanges(), and Range::overlap().

Referenced by astypes().

bool alternativeDonor	(	const Noschain &	ch1,
		const Noschain &	ch2,
		int &	i,
		int &	j,
		ASCountKeeper &	ack
	)

References ASCountKeeper::addAD(), Range::begin(), Range::end(), Range::isPlus(), Noschain::numberOfRanges(), and Range::overlap().

Referenced by astypes().

int astypes	(	const Noschain &	ch1,
		const Noschain &	ch2,
		int	tc[12],
		map< int, int > &	ritype,
		ASCountKeeper &	ack,
		map< int, int > &	ril,
		map< int, int > &	nril,
		set< pair< int, int > > &	majorRI,
		set< pair< int, int > > &	minorRI
	)

Parameters:

	tc	type of alternative splicing count, this function update this counter, each element represent a different type of AS. 0 AD, 1, AA, 2. CE, 3, CE_AD 4. CE_AA, 5. CE_ADAA, 6 IR 7. ADAA, 8 CEorAAend, 9 MutualExlusive, 10 other ASTYPE[10] ={ "AD", "AA", "CE", "CE_AD", "CE_AA", "CE_ADAA", "IR", "CEorAAend", "Other" };
	ch1	is the representative model and usually is the star model with the hightest EST count. For certain AS types, we may want to distinguish this *model from other alternative models. This models can be considered as the constitutive. There is a directionality between ch1 and ch2
	ch2	is usually the minor isoforms.
	ril	Retained Intron Length => Count

Returns:

the total number of AS found for this pair of models.

References adskip(), alternativeAcceptor(), alternativeDonor(), before(), bothADAA(), CEorAA(), Range::length(), length, mutualExclusive(), Noschain::numberOfRanges(), retainIntron(), skipExon(), and skipExonVariant().

Referenced by statAStype().

bool before	(	const Noschain &	ch1,
		const Noschain &	ch2,
		const int	i,
		const int	j
	)

References Range::begin(), and Range::direction().

Referenced by astypes(), mutualExclusive(), and skipExonVariant().

bool bothADAA	(	const Noschain &	ch1,
		const Noschain &	ch2,
		int &	i,
		int &	j,
		ASCountKeeper &	ack
	)

will increment i and j if success

References ASCountKeeper::ADAA, Range::begin(), Range::direction(), Range::end(), Range::isPlus(), and Noschain::numberOfRanges().

Referenced by astypes().

bool CEorAA	(	const Noschain &	ch1,
		const Noschain &	ch2,
		int &	i,
		int &	j,
		ASCountKeeper &	ack
	)

References Range::begin(), ASCountKeeper::CEorAAend, Range::end(), Range::isPlus(), and Noschain::numberOfRanges().

Referenced by astypes().

int main	(	int	argc,
		char *	argv[]
	)

References MysqlDBInfo::getAuthenInfo(), MysqlDBInfo::getPassword(), MysqlDBInfo::getUser(), statAStype(), and usage.

bool mutualExclusive	(	const Noschain &	ch1,
		const Noschain &	ch2,
		int &	i,
		int &	j,
		ASCountKeeper &	ack
	)

only check mutual exclusive of two exons, should check more

References ASCountKeeper::addME(), before(), Range::begin(), Range::end(), Noschain::numberOfRanges(), and Range::overlapFraction().

Referenced by astypes().

int retainIntron	(	const Noschain &	lch,
		const Noschain &	sch,
		int &	i,
		int &	j,
		ASCountKeeper &	ack,
		set< pair< int, int > > &	retained
	)

Returns:

the number of introns retained

Parameters:

	lch	is the chain that has introns retained.
	sch	is the one with introns spliced out. The two input chains must be in the same direction.
	retained	container for retained intron bounds. This container will be used as accumulator.

References Range::begin(), Range::direction(), Range::end(), ASCountKeeper::IR, and Noschain::numberOfRanges().

Referenced by astypes().

int skipExon	(	const Noschain &	lch,
		const Noschain &	sch,
		int &	i,
		int &	j,
		ASCountKeeper &	ack
	)

this could measure skipping of multiple exons will also accumulate special kind of CE to Other, but will not return true.

Returns:

0 for no AS of any kind. 1 for CE 2 for special kind of CE, group to Other 3 for even more strange kind.

References ASCountKeeper::addCE(), ASCountKeeper::addOther(), Range::begin(), Range::direction(), Range::end(), and Noschain::numberOfRanges().

Referenced by astypes().

int skipExonVariant	(	const Noschain &	ch1,
		const Noschain &	ch2,
		int &	i,
		int &	j,
		ASCountKeeper &	ack
	)

Returns:

1 for CE+AD, 2 for CE+AA, 3 for CE+AP(both AD and AA) 0 for Pure CE, -1 for no CE of any kind. 2516987->2517044 2517755->2518074 2518358->2518489 2518696->2522344 2516987->2517044 2517755->2518071 2518694->2522344

References ASCountKeeper::addCE(), before(), Range::begin(), Range::end(), Range::isPlus(), Noschain::numberOfRanges(), and Range::overlap().

Referenced by astypes().

void statAStype	(	Connection &	conn,
		const string &	dbname,
		const string &	resfile,
		bool	use_raw
	)

This is the main function. The input is taken from combest_good. I should have an option to take from combest_model/combest_gene.

References ASCountKeeper::accumulate(), ASTYPE, and astypes().

Referenced by main().

void usage

(

)

given gmap summary format, this program converts it into combest archive format (*.car)

It can be used a pipe, or given specific file names.

---

Variable Documentation

const char* ASTYPE[12] = {"AD", "AA", "CE", "CE_AD", "CE_AA", "CE_ADAA", "IR", "ADAA", "CEorAAend", "ME", "Other", ""}

Description This program will count the different categories of AS (currently I have 10 different kinds) by comparing the rep model of each gene to minor models. The rep model is mainly selected by the max number of ESTs covering this model plus using some auxiliary information such as CDS length, strange UTR feature. This is done by gathergene program that contains the detailed algorithm.

In addition this program focused on RI types. It produces the coordinates of Major and Minor RI introns and write them into different files. To get non-RI introns you need to do a set operation All_Introns - (major + minor) = nonRI introns. This is not done in this program. intronstop program computes the coordiantes of all introns and tells you how many Ns are inside.

Referenced by statAStype().