ESTAssembly Class Reference

#include <RNAModel.h>

Inheritance diagram for ESTAssembly:

List of all members.

Public Member Functions
	ESTAssembly ()
	ESTAssembly (const Noschain &seg, int nume, Coverdepth *bc, const string &chrom, const string &gs, int cngid)
	ESTAssembly (const Noschain &seg, int gcdsb_, int gcdse_, int nume, Coverdepth *bc, const string &chrom, const string &gs, int cngid, int fr)
	ESTAssembly (const Noschain &seg, int nume, const string &chrom, const string &gs, int cngid)
	ESTAssembly (const string &exons, const string &genomicid, const string &genomicseq, int modelid, int gcdsB, int gcdsE, int congregationid, int estcount, int gnid, int phase, float prof2star)
	ESTAssembly (const ESTAssembly &ea)
	~ESTAssembly ()
void	setNameAndTitle ()
void	setName ()
void	setTitle ()
void	setTitle (const string &prefix)
void	addTitleTag (const string &tag)
ESTAssembly &	operator= (const ESTAssembly &ea)
const string &	name () const
void	write (ostream &mod, ostream &exon, ostream &track, ostream &ousest, ostream &ouspep, char sep='\t') const
void	writetab (ostream &mod, ostream &exon, ostream &track, ostream &ousest, ostream &ouspep, char sep='\t') const
void	writeModel (ostream &ous, char sep='\t') const
float	RNACodingFraction ()
ESTAssembly *	breakSuffixModel () throw (PointOutChain)
ESTAssembly *	breakPrefixModel () throw (PointOutChain)
ESTAssembly *	prune3PrimeUTR () throw (PointOutChain)
ESTAssembly *	prune5PrimeUTR () throw (PointOutChain)
bool	shouldBreakSuffix () const
bool	shouldBreakPrefix () const
bool	isChimera () const
modeltype	objtype () const
ostream &	show (ostream &ous) const
ostream &	print (ostream &ous) const
void	showBaseCoverInfo (ostream &ous=cerr) const
int	ESTCount () const
int	numberOfEST () const
bool	validProfile () const
void	addComment (const string &comm)
void	writeCommentTab (ostream &ous) const
bool	hasComment () const
bool	noComment () const
void	fixIntronBound ()
int	intronFixState () const
void	checkIntronBound ()
bool	genuine () const
bool	semiGenuine () const
bool	isStar () const
bool	findDip (int b, int e, Dip &dip) const
ostream &	showProfile (ostream &ous) const
float	averageProfileHeight () const
int	maxProfileHeight () const
pair< float, float >	relativeProfileHeight (const ESTAssembly &ea) const
void	assignRPH (const ESTAssembly &star)
float	getRelprofh () const
void	setRelprofh (const float rh)
void	releaseProfile ()
ESTAssembly &	reverse (int RNACDSb, int RNACDSe)
Static Public Member Functions
static const char *	JGIModelColumns ()
static const char *	JGITranscriptColumns ()
static const char *	JGIProteinColumns ()
static bool	ORFLongEnough (pair< int, int > &bound, int ORFType, int trlen)
Static Public Attributes
static const char	modelheader [] = "modelid\tgeneid\tcongid\tgenomicid\tbegin\tend\tgCDSb\tgCDSe\tnumest\tnumexon\texonLength\texons\tCDSstart\tCDSend\tRNAseq\tpepseq\tframe\tprofileArea\tprofileMaxHeight\trelProfH\tbasecover"
static const char	trackheader [] = "id\tname\tchrom\tstrand\tstart\tend\tsfCount\tsfStarts\tsfEnds"
Protected Member Functions
ESTAssembly *	budPlusSuffixModel (pair< int, int > gcbts, int rcdsb, int rcdse, int suffix_gcdsb, int suffix_gcdse, const string &pep, int rcut1, int rcut2) throw (PointOutChain)
ESTAssembly *	budPlusSuffixModel (pair< int, int > gcuts, int rcdsb, int rcdse, int suffix_gcdsb, int suffix_gcdse, const string &pep) throw (PointOutChain)
ESTAssembly *	budMinusSuffixModel (pair< int, int > gcuts, int suffix_gcdsb, int suffix_gcdse, const string &subRNA, const string &pep, int usefull, pair< int, int > subcds, int rcut1, int rcut2) throw (PointOutChain)
ESTAssembly *	budMinusSuffixModel (pair< int, int > gcuts, int suffix_gcdsb, int suffix_gcdse, const string &subRNA, const string &pep, int usefull, pair< int, int > subcds) throw (PointOutChain)
ESTAssembly *	budPlusPrefixModel (pair< int, int > gcuts, int rcut1, int rcut2, int prefix_gcdsb, int prefix_gcdse, pair< int, int > subcds, int usefull, const string &maxPep) throw (PointOutChain)
ESTAssembly *	budMinusPrefixModel (pair< int, int > gcuts, int rcut1, int rcut2, int prefix_gcdsb, int prefix_gcdse, pair< int, int > subcds, const string &pep, const string &subRNA) throw (PointOutChain)
ESTAssembly *	budTemplate () const
void	resetNumest (ESTAssembly *mod)
void	resetSuffixProfile (ESTAssembly *mod, pair< int, int > gcuts)
void	resetPrefixProfile (ESTAssembly *mod, pair< int, int > gcuts)
void	setCDSInfo ()
Protected Attributes
int	congid
int	numest
Coverdepth *	basecovdep
float	relprofh
string	estname
string	esttitle
string	pepname
string	peptitle
string	comment
int	intronfix
Static Protected Attributes
static int	chimera_numestcut = 50
static int	chimera_peplencut = 90
static int	chimera_partial_peplencut = 40
static int	orfspace = 100

---

Detailed Description

this class is for replacement of ESTModel. This one is the driving force.

---

Constructor & Destructor Documentation

ESTAssembly::ESTAssembly

(

)

[inline]

default

Referenced by budTemplate().

ESTAssembly::ESTAssembly	(	const Noschain &	seg,
		int	nume,
		Coverdepth *	bc,
		const string &	chrom,
		const string &	gs,
		int	cngid
	)			[explicit]

used by Esambl to add Base cover depth info. This is currently used production version.

Parameters:

bc

is the base coverage profile pointer, it should be allocated on the free store. This object will manage to deallocate the memory.

References basecovdep, Range::direction(), fixIntronBound(), setNameAndTitle(), Coverdepth::setStrand(), and Coverdepth::strand().

ESTAssembly::ESTAssembly	(	const Noschain &	seg,
		int	gcdsb_,
		int	gcdse_,
		int	nume,
		Coverdepth *	bc,
		const string &	chrom,
		const string &	gs,
		int	cngid,
		int	fr
	)			[explicit]

used by breakup to make new models

References basecovdep, Range::direction(), fixIntronBound(), setNameAndTitle(), Coverdepth::setStrand(), and Coverdepth::strand().

ESTAssembly::ESTAssembly	(	const Noschain &	seg,
		int	nume,
		const string &	chrom,
		const string &	gs,
		int	cngid
	)			[explicit]

use without assigning the Coverdepth information faster, for single EST Assemblies where base coverage is not important.

References anyToString(), mRNAModel::cdsb, mRNAModel::cdse, estname, esttitle, RNAModel::genomicId(), RNAModel::getOid(), numest, pepname, peptitle, and Noschain::toString().

ESTAssembly::ESTAssembly	(	const string &	exons,
		const string &	genomicid,
		const string &	genomicseq,
		int	modelid,
		int	gcdsB,
		int	gcdsE,
		int	congregationid,
		int	estcount,
		int	gnid,
		int	phase,
		float	prof2star
	)			[inline]

for reading ESTAssembly stored in files

Parameters:

	modelid	Modelid is oid.
	gcdsB	and gcdsE are used to construct cdsb and cdse respectively. is geneid

References setNameAndTitle().

ESTAssembly::ESTAssembly

(

const ESTAssembly &

ea

)

[inline]

copy constructor problem with profile, Not making copy of the profile which is big. Only cp the pointer. The old object should call releaseProfile()

ESTAssembly::~ESTAssembly

(

)

[inline]

the assembler produced the profile

References basecovdep.

---

Member Function Documentation

void ESTAssembly::setNameAndTitle

(

)

helper function use to make a human readable name and title from the numbers already stored in the object.

References setName(), and setTitle().

Referenced by ESTAssembly().

void ESTAssembly::setName

(

)

The name is CE<oid>_<numest> for Model CEP<oid>_<numest> for protein.

References anyToString(), estname, RNAModel::getOid(), numest, and pepname.

Referenced by budMinusPrefixModel(), budMinusSuffixModel(), budPlusPrefixModel(), budPlusSuffixModel(), prune3PrimeUTR(), prune5PrimeUTR(), and setNameAndTitle().

void ESTAssembly::setTitle

(

)

References anyToString(), mRNAModel::cdsb, mRNAModel::cdse, esttitle, RNAModel::genomicId(), numest, and peptitle.

Referenced by budMinusPrefixModel(), budMinusSuffixModel(), budPlusPrefixModel(), budPlusSuffixModel(), prune3PrimeUTR(), prune5PrimeUTR(), and setNameAndTitle().

void ESTAssembly::setTitle

(

const string &

prefix

)

References anyToString(), mRNAModel::cdsb, mRNAModel::cdse, esttitle, RNAModel::genomicId(), numest, and peptitle.

void ESTAssembly::addTitleTag

(

const string &

tag

)

References esttitle, and peptitle.

Referenced by ESTAssemblyid::breakup().

ESTAssembly & ESTAssembly::operator=

(

const ESTAssembly &

ea

)

References comment, congid, estname, esttitle, intronfix, numest, mRNAModel::operator=(), pepname, peptitle, and relprofh.

const string& ESTAssembly::name

(

)

const [inline]

References estname.

void ESTAssembly::write	(	ostream &	mod,
		ostream &	exon,
		ostream &	track,
		ostream &	ousest,
		ostream &	ouspep,
		char	sep = '\t'
	)			const

when this methods is called, the id, and exonid will be incremented Produces output in two different formats.

References Range::begin(), mRNAModel::cdsb, mRNAModel::cdse, congid, Range::end(), estname, esttitle, RNAModel::exonLength(), mRNAModel::gcdsb, mRNAModel::gcdse, RNAModel::genomicId(), RNAModel::getOid(), mRNAModel::getProtein(), Noschain::jgiformat(), Noschain::numberOfRanges(), numest, mRNAModel::pep, pepname, peptitle, printFasta(), RNAModel::rna, RNAModel::RNASequence(), Noschain::toString(), and RNAModel::writeExon().

void ESTAssembly::writetab	(	ostream &	mod,
		ostream &	exon,
		ostream &	track,
		ostream &	ousest,
		ostream &	ouspep,
		char	sep = '\t'
	)			const

Parameters:

	mod	combest tabular format with the following columns: objectid, geneid, congregationid, genomicid, begin, end, genomic_cdsb, genomic_cdse, numest, numexon, sumexonlength, exon_structure, mRNAcdsb, mRNAcdse, RNAseq, pepseq, frame, area_of_profile, maxheight_profile, relprofh2star, basecover this output used writeModel() Not all are calculated, for single EST model we don't calculate the area or do't generate the base profile.
	exon	exontable columns: exonid, model_objectid, exb, exe. sequencial from first to last exon
	track.	JGI model row
	ousest	JGI transcript table row
	ouspep	JGI protein table row

This function writes a lots of redundant information to different files. This is for convinience. For performance, a minimal output file should be generated.

Reimplemented from mRNAModel.

Reimplemented in ESTAssemblyid.

References comment, estname, esttitle, mRNAModel::JGIModelRow(), mRNAModel::JGIProteinRow(), mRNAModel::JGITranscriptRow(), pepname, peptitle, RNAModel::writeExon(), and writeModel().

void ESTAssembly::writeModel	(	ostream &	ous,
		char	sep = '\t'
	)			const

write the model in tabular format for storage. This method does not inherit from the parent classes. This method stores all essential information and some summary information for quick SQL programming. for relative profile height to star, only the minimum is outputed. For 99.9% models, maximum should be 1. The columns: ================================================= oid, GeneId, congid, genomicId(), begin, end, gcdsb, gcdse, numest, numberOfRanges, sumExonLength, exonsStringFormat[b1-e1,b2-e2,...], cdsb, cdse, rnaSeq, pep, frame, BCParea, BCPmaxh, BCPrelheight, actualBCP =================================================

exonStringFormat use the toString() method of Noschain.

Reimplemented in ESTAssemblyid.

References basecovdep, Range::begin(), mRNAModel::cdsb, mRNAModel::cdse, congid, Range::end(), RNAModel::exonLength(), mRNAModel::frame, mRNAModel::gcdsb, mRNAModel::gcdse, RNAModel::genomicId(), RNAModel::getGeneId(), RNAModel::getOid(), Coverdepth::maxAndSum(), Noschain::numberOfRanges(), numest, mRNAModel::pep, relprofh, RNAModel::rna, RNAModel::RNALength(), and Noschain::toString().

Referenced by ESTAssemblyid::writeModel(), and writetab().

float ESTAssembly::RNACodingFraction

(

)

[inline]

References mRNAModel::CDSLength(), and RNAModel::RNALength().

ESTAssembly * ESTAssembly::breakSuffixModel

(

)

throw (PointOutChain)

Break off model contained inside the 3'-UTR if 1. the protein is complete, 2. number of EST > 100 3. 5'-UTR or 3'-UTR exonLength() > 700 4. UTR has at least one intron then make the UTR a new Model These new models may be identical from different genes, so some set operation needs to be performed. How to break off the new UTR from the main gene? This is an very expensive operation, slows down the program by a factor of 2.

When the suffix model has no intron, the algorithm should try ORF in both directions. Right now it is only testing in one direction. This is apparently not sufficient.

Returns:

a pointer to a new ESTAssembly if success return 0 if cannot break off new model.

Reimplemented in ESTAssemblyid.

References addComment(), Dip::breakPoint(), budMinusSuffixModel(), budPlusSuffixModel(), mRNAModel::cdse, Range::direction(), Range::end(), esttitle, findDip(), mRNAModel::gcdse, RNAModel::genomicIndex(), RNAModel::getOid(), itos(), longestNoStartORFPlus(), longestNoStopORFPlus(), maxFullORF(), Noschain::numberOfRanges(), ORFLongEnough(), prune3PrimeUTR(), reverseComplement(), RNAModel::rna, shouldBreakSuffix(), show(), mRNAModel::ThreePrimeUTR(), mRNAModel::ThreePrimeUTRLength(), and OutsideGenomicSequence::what().

Referenced by breakExtraModel(), and ESTAssemblyid::breakSuffixModel().

ESTAssembly * ESTAssembly::breakPrefixModel

(

)

throw (PointOutChain)

analogous to breakSuffixModel()

Reimplemented in ESTAssemblyid.

References addComment(), Range::begin(), Dip::breakPoint(), budMinusPrefixModel(), budPlusPrefixModel(), mRNAModel::cdsb, comment, Range::direction(), findDip(), mRNAModel::FivePrimeUTR(), mRNAModel::gcdsb, RNAModel::genomicIndex(), RNAModel::getOid(), itos(), longestNoStartORFPlus(), longestNoStopORFPlus(), maxFullORF(), Noschain::numberOfRanges(), ORFLongEnough(), prune5PrimeUTR(), reverseComplement(), RNAModel::rna, RNAModel::RNAIndex(), shouldBreakPrefix(), show(), and OutsideGenomicSequence::what().

Referenced by breakExtraModel(), and ESTAssemblyid::breakPrefixModel().

ESTAssembly * ESTAssembly::prune3PrimeUTR

(

)

throw (PointOutChain)

this function does not compute ORF first, but simply look for a Dip between orfspace nt away from either end of ORF to the limits of the genomic DNA. if found it will prune off the extra Assembly, then it will try to set up coding info for this new Assembly This is apposed to breakPrefix or breakSuffix methods where they first compute ORF then try to cut the assembly between the ORFs.

parent

References addComment(), Noschain::advancePosOnExon(), Dip::breakPoint(), budTemplate(), Range::direction(), Range::end(), findDip(), mRNAModel::gcdse, Range::length(), mRNAModel::num3NoncodingExons(), orfspace, resetNumest(), resetSuffixProfile(), RNAModel::rna, RNAModel::RNAIndex(), setCDSInfo(), Noschain::setChain(), setName(), setTitle(), show(), Noschain::subchain(), mRNAModel::ThreePrimeUTRLength(), mRNAModel::trimAfterPoint(), and mRNAModel::valid().

Referenced by breakSuffixModel().

ESTAssembly * ESTAssembly::prune5PrimeUTR

(

)

throw (PointOutChain)

References addComment(), Range::begin(), Dip::breakPoint(), budTemplate(), Range::direction(), findDip(), mRNAModel::FivePrimeUTRLength(), mRNAModel::gcdsb, Range::length(), mRNAModel::num5NoncodingExons(), orfspace, resetNumest(), resetPrefixProfile(), Noschain::retreatPosOnExon(), RNAModel::rna, RNAModel::RNAIndex(), setCDSInfo(), Noschain::setChain(), setName(), setTitle(), show(), Noschain::subchain(), mRNAModel::trimBeforePoint(), and mRNAModel::valid().

Referenced by breakPrefixModel().

bool ESTAssembly::shouldBreakSuffix

(

)

const

if UTR length > 1500 and has stop then break

References chimera_numestcut, mRNAModel::hasStop(), Noschain::numberOfRanges(), numest, mRNAModel::ThreePrimeUTR(), and mRNAModel::ThreePrimeUTRLength().

Referenced by breakSuffixModel().

bool ESTAssembly::shouldBreakPrefix

(

)

const

References chimera_numestcut, mRNAModel::FivePrimeUTR(), mRNAModel::FivePrimeUTRLength(), mRNAModel::hasStart(), Noschain::numberOfRanges(), and numest.

Referenced by breakPrefixModel().

bool ESTAssembly::isChimera

(

)

const

This model is more likely to be Chimera

fungal UTRs could be short so we are setting a 500 nt limit

References mRNAModel::CDSFractionRNA(), chimera_numestcut, mRNAModel::FivePrimeUTR(), mRNAModel::FivePrimeUTRLength(), Range::length(), Noschain::numberOfRanges(), numest, mRNAModel::ThreePrimeUTR(), and mRNAModel::ThreePrimeUTRLength().

Referenced by Graphid::assemble(), and testBreakup().

modeltype ESTAssembly::objtype

(

)

const [inline, virtual]

Reimplemented from mRNAModel.

References EST.

ostream & ESTAssembly::show

(

ostream &

ous

)

const [virtual]

ESTAssembly for debug

Reimplemented from mRNAModel.

References congid, esttitle, numest, peptitle, and mRNAModel::show().

Referenced by breakPrefixModel(), breakSuffixModel(), findDip(), prune3PrimeUTR(), prune5PrimeUTR(), readESTModel(), and updateOneUpdated().

ostream & ESTAssembly::print

(

ostream &

ous

)

const [virtual]

for operator<<

Reimplemented from mRNAModel.

References congid, esttitle, numest, peptitle, and mRNAModel::print().

void ESTAssembly::showBaseCoverInfo

(

ostream &

ous = cerr

)

const [inline]

References basecovdep.

int ESTAssembly::ESTCount

(

)

const [inline]

References numest.

int ESTAssembly::numberOfEST

(

)

const [inline]

References numest.

Referenced by mRNAModelUpdate::addESTCover().

bool ESTAssembly::validProfile

(

)

const

References basecovdep, Range::direction(), Coverdepth::exons(), Noschain::reverse(), and Coverdepth::strand().

void ESTAssembly::addComment

(

const string &

comm

)

References comment.

Referenced by breakPrefixModel(), breakSuffixModel(), prune3PrimeUTR(), and prune5PrimeUTR().

void ESTAssembly::writeCommentTab

(

ostream &

ous

)

const [inline]

References comment, and RNAModel::getOid().

bool ESTAssembly::hasComment

(

)

const [inline]

References comment.

bool ESTAssembly::noComment

(

)

const [inline]

References comment.

void ESTAssembly::fixIntronBound

(

)

Returns:

a number 0 for good intron bounds for all introns. 1 fixed at least one 2 bad intron bound of at least one intron 3 fixed at least one intron, but there are still at least one bad intron bounds remaining.

bad intron bounds are the ones other than GT..AG, GC..AG, AT..AC Will also modify inton boundary of the profile. Will erase exon and fill intron part with whatever profile value near the exon/inton boundary.

NNN is considered good.

Only fix introngs longer than 32 nt. I left short ones intact in case there are RNA-editing enzymes. Also make sure exons are long enough so that it will not be chaged into different directions. So far I am 4 down 5 up 9 nt around the splice sites. 4 exon, 5 intron bases.

References basecovdep, Range::begin(), Range::direction(), Range::end(), Coverdepth::erase(), Noschain::exons, Coverdepth::extendLeft(), Coverdepth::extendRight(), RNAModel::gseq, intronfix, Noschain::numberOfRanges(), and mRNAModel::reset().

Referenced by ESTAssembly().

int ESTAssembly::intronFixState

(

)

const [inline]

References intronfix.

Referenced by assembleChain().

void ESTAssembly::checkIntronBound

(

)

this function will be used by breaking family of methods. If the model has been broken, I will have to recheck to see if there is a bad intron IN this case it will only know two stats, good (0) bad(2), it does not knonw whether it got fixed or not. Because there is a tiny fraction of models are like this, this is fine.

This method will set the internal state variable: intronfix This method coulb potentially produce bad exons of inverted direction!.

References Range::begin(), Range::direction(), Range::end(), Noschain::exons, RNAModel::gseq, intronfix, and Noschain::numberOfRanges().

Referenced by breakExtraModel().

bool ESTAssembly::genuine

(

)

const [inline]

use mRNAModel genuine method

Reimplemented from mRNAModel.

References mRNAModel::genuine(), and intronfix.

Referenced by assembleChain().

bool ESTAssembly::semiGenuine

(

)

const [inline]

Reimplemented from mRNAModel.

References intronfix, and mRNAModel::semiGenuine().

bool ESTAssembly::isStar

(

)

const [inline]

good quality model: UTR < 600 nt, CDS length > 330 nt, numUTR exon < 2 or CDS fraction > 65%

Reimplemented from mRNAModel.

References intronfix, and mRNAModel::isStar().

bool ESTAssembly::findDip	(	int	b,
		int	e,
		Dip &	dip
	)			const

[b,e] genomic position Given a range [b,e], this function will try to find a Dip

Returns:

true if found one, false else.

Parameters:

dip

the value of dip will be set to the result.

References basecovdep, Range::begin(), Range::end(), Range::length(), Noschain::numberOfRanges(), Coverdepth::searchDip(), show(), Noschain::subchain(), and PointOutChain::what().

Referenced by breakPrefixModel(), breakSuffixModel(), ESTAssemblyid::breakup(), prune3PrimeUTR(), and prune5PrimeUTR().

ostream& ESTAssembly::showProfile

(

ostream &

ous

)

const [inline]

display the profile

References basecovdep.

float ESTAssembly::averageProfileHeight

(

)

const [inline]

This value is not veray useful

References Coverdepth::averageHeight(), and basecovdep.

int ESTAssembly::maxProfileHeight

(

)

const [inline]

References basecovdep, and Coverdepth::maxHeight().

pair<float,float> ESTAssembly::relativeProfileHeight

(

const ESTAssembly &

ea

)

const [inline]

this is a simple wrapper for Coverdepth object relativeHeight()

Returns:

(min,max) profile height ratio of ea/this Assembly

References basecovdep, and Coverdepth::relativeHeight().

Referenced by assignRPH().

void ESTAssembly::assignRPH

(

const ESTAssembly &

star

)

[inline]

This method looks at thing from the opposite point of view as relativeProfileHeight. this/star star is used as reference. relative profile height of -1 means the star model and this models has no common profile.

References relativeProfileHeight(), and relprofh.

float ESTAssembly::getRelprofh

(

)

const [inline]

References relprofh.

void ESTAssembly::setRelprofh

(

const float

rh

)

[inline]

References relprofh.

void ESTAssembly::releaseProfile

(

)

[inline]

will not destroy the profile, simply transfer ownership. This is called after giving away the profile.

References basecovdep.

Referenced by ESTAssemblyid::ESTAssemblyid().

ESTAssembly& ESTAssembly::reverse	(	int	newRNACDSB,
		int	newRNACDSE
	)			[inline]

prevents reversing of multiexon models. it will crash, if you try to. This will prevent the programmer from making mistakes. cds range, pep, and frame will all needs to be reset.

Reimplemented from mRNAModel.

References basecovdep, RNAModel::reverse(), and Coverdepth::reverseStrand().

const char * ESTAssembly::JGIModelColumns

(

)

[static]

use header[] to construct these headers from base class information

References mRNAModel::header, and mRNAModel::jgiModelCol.

Referenced by makeModels().

const char * ESTAssembly::JGITranscriptColumns

(

)

[static]

References mRNAModel::header, and mRNAModel::jgiTranscriptCol.

Referenced by makeModels().

const char * ESTAssembly::JGIProteinColumns

(

)

[static]

References mRNAModel::header, and mRNAModel::jgiProteinCol.

Referenced by makeModels().

bool ESTAssembly::ORFLongEnough	(	pair< int, int > &	bound,
		int	ORFType,
		int	trlen
	)			[static]

Parameters:

	ORFType	[0,1,2,-1]
	trlen	length of the parent transcript 1 nostart or nostop. M--- or ----* 2 full M---* 0 nostart and nostop ---- -1 nothing, this is not possible as long as RNA longer than 3 nt, you should get something.

References chimera_partial_peplencut, and chimera_peplencut.

Referenced by breakPrefixModel(), and breakSuffixModel().

ESTAssembly * ESTAssembly::budPlusSuffixModel	(	pair< int, int >	gcbts,
		int	rcdsb,
		int	rcdse,
		int	suffix_gcdsb,
		int	suffix_gcdse,
		const string &	pep,
		int	rcut1,
		int	rcut2
	)			throw (PointOutChain) [protected]

helper function for Break Suffix Model

References budTemplate(), mRNAModel::cdsb, mRNAModel::cdse, Range::end(), mRNAModel::gcdsb, mRNAModel::gcdse, RNAModel::getOid(), itos(), mRNAModel::pep, resetNumest(), resetSuffixProfile(), RNAModel::rna, Noschain::setChain(), setName(), mRNAModel::setProtein(), setTitle(), Noschain::subchain(), and mRNAModel::trimAfterPoint().

Referenced by breakSuffixModel(), and budPlusSuffixModel().

ESTAssembly* ESTAssembly::budPlusSuffixModel	(	pair< int, int >	gcuts,
		int	rcdsb,
		int	rcdse,
		int	suffix_gcdsb,
		int	suffix_gcdse,
		const string &	pep
	)			throw (PointOutChain) [inline, protected]

shorter version of above function

References budPlusSuffixModel(), mRNAModel::pep, and RNAModel::RNAIndex().

ESTAssembly * ESTAssembly::budMinusSuffixModel	(	pair< int, int >	gcuts,
		int	suffix_gcdsb,
		int	suffix_gcdse,
		const string &	subRNA,
		const string &	pep,
		int	usefull,
		pair< int, int >	subcds,
		int	rcut1,
		int	rcut2
	)			throw (PointOutChain) [protected]

full parameter version

References budTemplate(), mRNAModel::cdsb, mRNAModel::cdse, Range::end(), mRNAModel::frame, mRNAModel::gcdsb, mRNAModel::gcdse, RNAModel::getOid(), itos(), mRNAModel::pep, resetNumest(), resetSuffixProfile(), RNAModel::rna, Noschain::setChain(), setName(), mRNAModel::setProtein(), RNAModel::setRNA(), setTitle(), Noschain::subchain(), and mRNAModel::trimAfterPoint().

Referenced by breakSuffixModel(), and budMinusSuffixModel().

ESTAssembly* ESTAssembly::budMinusSuffixModel	(	pair< int, int >	gcuts,
		int	suffix_gcdsb,
		int	suffix_gcdse,
		const string &	subRNA,
		const string &	pep,
		int	usefull,
		pair< int, int >	subcds
	)			throw (PointOutChain) [inline, protected]

shorter version

References budMinusSuffixModel(), mRNAModel::pep, and RNAModel::RNAIndex().

ESTAssembly * ESTAssembly::budPlusPrefixModel	(	pair< int, int >	gcuts,
		int	rcut1,
		int	rcut2,
		int	prefix_gcdsb,
		int	prefix_gcdse,
		pair< int, int >	subcds,
		int	usefull,
		const string &	maxPep
	)			throw (PointOutChain) [protected]

helper function

References Range::begin(), budTemplate(), mRNAModel::cdsb, mRNAModel::cdse, mRNAModel::frame, mRNAModel::gcdsb, mRNAModel::gcdse, RNAModel::getOid(), itos(), resetNumest(), resetPrefixProfile(), RNAModel::rna, Noschain::setChain(), setName(), mRNAModel::setProtein(), setTitle(), Noschain::subchain(), and mRNAModel::trimBeforePoint().

Referenced by breakPrefixModel().

ESTAssembly * ESTAssembly::budMinusPrefixModel	(	pair< int, int >	gcuts,
		int	rcut1,
		int	rcut2,
		int	prefix_gcdsb,
		int	prefix_gcdse,
		pair< int, int >	subcds,
		const string &	pep,
		const string &	subRNA
	)			throw (PointOutChain) [protected]

Parameters:

subRNA

is the parent rna.substr(0,cdsb-1) reverse complemented

References Range::begin(), budTemplate(), mRNAModel::cdsb, mRNAModel::cdse, mRNAModel::gcdsb, mRNAModel::gcdse, RNAModel::getOid(), itos(), mRNAModel::pep, resetNumest(), resetPrefixProfile(), Noschain::setChain(), setName(), mRNAModel::setProtein(), RNAModel::setRNA(), setTitle(), Noschain::subchain(), and mRNAModel::trimBeforePoint().

Referenced by breakPrefixModel().

ESTAssembly * ESTAssembly::budTemplate

(

)

const [protected]

produce a model with the invariant information: congid, gid, gseq

References congid, ESTAssembly(), RNAModel::gid, and RNAModel::gseq.

Referenced by budMinusPrefixModel(), budMinusSuffixModel(), budPlusPrefixModel(), budPlusSuffixModel(), prune3PrimeUTR(), and prune5PrimeUTR().

void ESTAssembly::resetNumest

(

ESTAssembly *

mod

)

[protected]

assume profile exists for this and mod

References Coverdepth::area(), basecovdep, max, and numest.

Referenced by budMinusPrefixModel(), budMinusSuffixModel(), budPlusPrefixModel(), budPlusSuffixModel(), prune3PrimeUTR(), and prune5PrimeUTR().

void ESTAssembly::resetSuffixProfile	(	ESTAssembly *	mod,
		pair< int, int >	gcuts
	)			[protected]

the gcuts.first is on parent side, gcuts.second in on suffix side

References basecovdep, Range::direction(), Range::end(), Coverdepth::subprofile(), Coverdepth::truncateHead(), and Coverdepth::truncateTail().

Referenced by budMinusSuffixModel(), budPlusSuffixModel(), and prune3PrimeUTR().

void ESTAssembly::resetPrefixProfile	(	ESTAssembly *	mod,
		pair< int, int >	gcuts
	)			[protected]

for prefix it is differet, gcuts.first is on prefix side

References basecovdep, Range::begin(), Range::direction(), Coverdepth::subprofile(), Coverdepth::truncateHead(), and Coverdepth::truncateTail().

Referenced by budMinusPrefixModel(), budPlusPrefixModel(), and prune5PrimeUTR().

void ESTAssembly::setCDSInfo

(

)

[protected]

this function is a helper for breaking chimera EST assemblies It requires that both RNA sequences has just been set and missing protein and CDS range information. If the gene is single exon, it will try both directions to find the max ORF. At the end it uses the CDS info at RNA level to set CDS info at genomic level.

References mRNAModel::cdsb, mRNAModel::cdse, mRNAModel::frame, mRNAModel::gcdsb, mRNAModel::gcdse, RNAModel::genomicIndex(), longestORFPlus(), Noschain::numberOfRanges(), mRNAModel::pep, Noschain::reverse(), reverseComplementInPlace(), RNAModel::rna, and Noschain::show().

Referenced by prune3PrimeUTR(), and prune5PrimeUTR().

---

Member Data Documentation

const char ESTAssembly::modelheader = "modelid\tgeneid\tcongid\tgenomicid\tbegin\tend\tgCDSb\tgCDSe\tnumest\tnumexon\texonLength\texons\tCDSstart\tCDSend\tRNAseq\tpepseq\tframe\tprofileArea\tprofileMaxHeight\trelProfH\tbasecover" [static]

Reimplemented from mRNAModel.

const char ESTAssembly::trackheader = "id\tname\tchrom\tstrand\tstart\tend\tsfCount\tsfStarts\tsfEnds" [static]

int ESTAssembly::congid [protected]

Referenced by ESTAssemblyid::breakup(), budTemplate(), operator=(), print(), show(), write(), and writeModel().

int ESTAssembly::numest [protected]

Referenced by ESTAssemblyid::breakup(), ESTAssembly(), ESTCount(), isChimera(), numberOfEST(), operator=(), print(), resetNumest(), setName(), setTitle(), shouldBreakPrefix(), shouldBreakSuffix(), show(), write(), and writeModel().

Coverdepth* ESTAssembly::basecovdep [protected]

base coverage depth, set to zero if no profile.

Referenced by averageProfileHeight(), ESTAssemblyid::breakup(), ESTAssembly(), findDip(), fixIntronBound(), maxProfileHeight(), relativeProfileHeight(), releaseProfile(), resetNumest(), resetPrefixProfile(), resetSuffixProfile(), reverse(), showBaseCoverInfo(), showProfile(), validProfile(), writeModel(), and ~ESTAssembly().

float ESTAssembly::relprofh [protected]

relative profile height. A convenient number for (min,max) profile height relative to the star model that is defined as the model with the most EST. If gene has only one model, then the value will be (1,1). Or if the model is star model, the value is (1,1). Actually storing two numbers is not needed. We will only store one number min(this/star) here star model is defined simply as model with max number of EST. If more than one, simply pick one randomly.

Referenced by assignRPH(), getRelprofh(), operator=(), setRelprofh(), and writeModel().

string ESTAssembly::estname [protected]

Referenced by ESTAssembly(), name(), operator=(), setName(), write(), ESTAssemblyid::writetab(), and writetab().

string ESTAssembly::esttitle [protected]

Referenced by addTitleTag(), breakSuffixModel(), ESTAssembly(), operator=(), print(), setTitle(), show(), write(), ESTAssemblyid::writetab(), and writetab().

string ESTAssembly::pepname [protected]

Referenced by ESTAssembly(), operator=(), setName(), write(), ESTAssemblyid::writetab(), and writetab().

string ESTAssembly::peptitle [protected]

Referenced by addTitleTag(), ESTAssembly(), operator=(), print(), setTitle(), show(), write(), ESTAssemblyid::writetab(), and writetab().

string ESTAssembly::comment [protected]

use to store information about frame-shift etc

Referenced by addComment(), breakPrefixModel(), ESTAssemblyid::breakup(), hasComment(), noComment(), operator=(), writeCommentTab(), and writetab().

int ESTAssembly::intronfix [protected]

a state variable to remember whether the inron bound has been fiexd or not. only less 0.5% of the models have bad intron bounds that are derive from either bad sequence or artifacts or true biological processes. 0 good, no fix one of the three known GT(C)..AG, AT..AC 1 fixed at least one intron 2 at least one bad bound, 3 fixed at least one, and still have bad intron bounds. The fixIntronBound() function sets this variable.

Referenced by checkIntronBound(), fixIntronBound(), genuine(), intronFixState(), isStar(), operator=(), and semiGenuine().

int ESTAssembly::chimera_numestcut = 50 [static, protected]

Referenced by isChimera(), shouldBreakPrefix(), and shouldBreakSuffix().

int ESTAssembly::chimera_peplencut = 90 [static, protected]

when ESt has joint two models like the following ---====--=======----- while trying to break chimera EST into individual models, we have a cutoff for the shortest peptide.

Referenced by ORFLongEnough().

int ESTAssembly::chimera_partial_peplencut = 40 [static, protected]

Referenced by ORFLongEnough().

int ESTAssembly::orfspace = 100 [static, protected]

Now I used it to trim 3' or 5' long UTRs Now I am setting it to 100 nt. It look for cutting point only 100 nt away from either ends of an UTR

Referenced by prune3PrimeUTR(), and prune5PrimeUTR().

---

The documentation for this class was generated from the following files:

• range/RNAModel.h
• range/RNAModel.cpp