RetroSearch Browse

)m.

().size(); ++

) {

458 if

(m.

()[

-1].m_ssplice_sig ==

"XX"

|| m.

()[

].m_fsplice_sig ==

"XX"

)

461 if

(

(lim_for_nested, range)) {

462 bool

is_hole = !m.

()[

-1].m_ssplice || !m.

()[

].m_fsplice;

463

TGenomeRange intron(range, is_hole, ig);

464

introns.Insert(intron);

471

list<TGIter> genes_hosting_partial;

472

list<TGIter> nested_partial;

473

GenomeRangeMap finished_intervals;

474 if

(!introns.empty()) {

475

list<TGIter> genes_to_remove;

476 for

(

auto

ig = genes.begin(); ig != genes.end(); ++ig) {

477

TIterList& modelsi = ig->second;

478 auto

gfront = modelsi.front();

481 if

(iclust != introns.end() &&

(iclust->first, lim_for_nested)) {

482 for

(TGenomeRange& intron : iclust->second) {

484 if

(

(range, lim_for_nested)) {

485 bool

is_hole = get<1>(intron);

486 auto

host_it = get<2>(intron);

487 if

(is_hole && !gfront->GoodEnoughToBeAnnotation()) {

488 if

(host_it->second.front()->Score() > gfront->Score())

489

genes_to_remove.push_back(ig);

491

genes_to_remove.push_back(host_it);

493 if

(gfront->GoodEnoughToBeAnnotation()) {

494 for

(

auto

im : modelsi)

497

genes_hosting_partial.push_back(host_it);

498

nested_partial.push_back(ig);

505 if

(gfront->GoodEnoughToBeAnnotation()) {

507 for

( ; !found && iclust != introns.end() && iclust->first.IntersectingWith(lim_for_nested); ++iclust) {

508 for

(TGenomeRange& intron : iclust->second) {

509 if

(get<2>(intron) == ig)

514

TGenomeRange finished_interval(lim_for_nested,

false

, ig);

515

finished_intervals.Insert(finished_interval);

522

genes_to_remove.sort(geneid_order());

523

genes_to_remove.unique();

524

genes_hosting_partial.sort(geneid_order());

525

genes_hosting_partial.unique();

526

nested_partial.sort(geneid_order());

527

nested_partial.unique();

528 for

(

auto

it : genes_to_remove) {

529

genes_hosting_partial.remove(it);

530

nested_partial.remove(it);

531 for

(

auto

im : it->second) {

533

im->AddComment(

"Partial gene in a hole"

);

534

bad_aligns.push_back(*im);

542

GenomeRangeMap hosting_intervals;

543 for

(

auto

it : genes_hosting_partial) {

544

TIterList& lst = it->second;

546 bool

coding_gene = find_if(lst.begin(), lst.end(), [](TGeneModelList::iterator im){ return im->ReadingFrame().NotEmpty(); }) != lst.end();

550 for

(

auto

im : lst) {

557

gene_lim_for_nested += model_lim_for_nested;

560

vector<int> grange(gene_lim_for_nested.

GetLength

(),1);

561 for

(

auto

im : lst) {

570 for

(

int i

= 0;

< (

)ai.

().size(); ++

) {

572 for

(

int

j = overlap.

(); j <= overlap.

(); ++j)

573

grange[j-gene_lim_for_nested.

()] = 0;

576 for

(

int i

= 1;

< (

)ai.

().size(); ++

) {

577 if

(!ai.

()[

-1].m_ssplice || !ai.

()[

].m_fsplice) {

580 for

(

int

j = hole.

(); j <= hole.

(); ++j)

581

grange[j-gene_lim_for_nested.

()] = 0;

585 _ASSERT

(grange.front() == 0 && grange.back() == 0);

589 for

(

int

j = 0; j < (

)grange.size(); ++j) {

591 if

(grange[j] == 1) {

595

}

else if

(grange[j] == 1) {

599

TGenomeRange hosting_interval(interval,

false

, it);

600

hosting_intervals.Insert(hosting_interval);

607

TRangeModels nested_models;

608 for

(

auto

ig : nested_partial) {

609

TGeneModelList::iterator nested_modeli = ig->second.front();

610 _ASSERT

(ig->second.size() == 1);

615 if

(rslt != hosting_intervals.end() &&

(rslt->first, lim_for_nested)) {

616 for

(

auto

& grange : rslt->second) {

618 if

(

(interval,lim_for_nested)) {

619 if

(hosting_interval.

Empty

())

620

hosting_interval = interval;

622

hosting_interval = (hosting_interval&interval);

628 if

(hosting_interval.

NotEmpty

()) {

629

TIterList nested(1,nested_modeli);

631 for

(

auto

it = finished_intervals.lower_bound(left); it != finished_intervals.end() && it->first.IntersectingWith(hosting_interval); ++it) {

632 for

(

auto

& grange : it->second) {

637 if

(

Precede

(finished_interval,lim_for_nested)) {

638

hosting_interval.

SetFrom

(finished_interval.

());

639

}

else if

(

Precede

(lim_for_nested,finished_interval)) {

640

hosting_interval.

SetTo

(finished_interval.

());

642 for

(

auto

im : get<2>(grange)->second)

643

nested.push_back(im);

648

nested_models[hosting_interval].splice(nested_models[hosting_interval].begin(), nested);

653 bool

scaffold_wall =

wall

;

655 ITERATE

(TRangeModels,

, nested_models) {

660 ITERATE

(TIterList, im,

->second) {

661

nested.push_back(**im);

663 if

(!(*im)->GoodEnoughToBeAnnotation()) {

664 if

(nested.back().HasStart() && !

(hosting_interval,nested.back().MaxCdsLimits())) {

665 CCDSInfo

cds = nested.back().GetCdsInfo();

666 if

(nested.back().Strand() ==

ePlus

)

670

nested.back().SetCdsInfo(cds);

672

nested.back().AddComment(

"partialnested"

);

675

included_complete_models.

insert

((*im)->ID());

679

cerr <<

"Interval "

<< hosting_interval <<

'\t'

<< nested.size() << endl;

684 if

(!im->Support().empty()) {

686 if

(im->ID() == 0 || included_complete_models.

find

(im->ID()) == included_complete_models.

end

())

687

models.push_back(*im);

691 wall

= scaffold_wall;

703 bool

gapfilled =

false

;

706 if

(ie->m_fsplice_sig ==

"XX"

|| ie->m_ssplice_sig ==

"XX"

)

709

genome_cds += (cds&ie->Limits()).

GetLength

();

712 if

(gapfilled && genome_cds < 45) {

714

model.

AddComment

(

"Most CDS in genomic gap"

);

715

bad_aligns.push_back(model);

727 if

(!models.empty()) {

728 for

(

auto

it_loop =

(models.begin()); it_loop != models.end(); ) {

730 if

(it->RankInGene() != 1 || it->GoodEnoughToBeAnnotation() || it->Type()&

CGeneModel::eNested

)

732 auto

it_prev =

(it);

733 if

(it_prev->RankInGene() != 1 || it_prev->GoodEnoughToBeAnnotation() || it_prev->Type()&

CGeneModel::eNested

)

736 if

(it->MaxCdsLimits().IntersectingWith(it_prev->MaxCdsLimits())) {

737

cerr <<

"Intersecting alignments "

<< it->ID() <<

" "

<< it_prev->ID() <<

" "

<< it->Score() <<

" "

<< it_prev->Score() << endl;

738 auto

it_erase = (it->Score() < it_prev->Score()) ? it : it_prev;

740

it_erase->AddComment(

"Intersects with other partial"

);

741

bad_aligns.push_back(*it_erase);

742

models.erase(it_erase);

755 Predict

(left, right, aligns.begin(), aligns.end(), models_tmp,(left!=0 ||

wall

, left!=0,

false

, bad_aligns);

757 if

(!it->Support().empty() || it->RealCdsLen() >=

minCdsLen

)

758

models.push_back(*it);

763 CCDSInfo

cds_info = it->GetCdsInfo();

769 if

(((

->IsInsertion() ||

->IsMismatch()) &&

(fullcds,

->Loc())) ||

770

(

->IsDeletion() &&

->Loc() > fullcds.

() &&

->Loc() <= fullcds.

())) {

774

it->FrameShifts() = fs;

783

it->SetCdsInfo(cds_info);

785 if

(it->PStop(

false

) || !it->FrameShifts().empty()) {

789 CCDSInfo

cds_info = it->GetCdsInfo();

791

it->SetCdsInfo(cds_info);

808 for

(five_p=0; five_p < (

)vec.size() && vec[five_p] ==

'N'

; ++five_p);

809 for

(three_p=0; three_p < (

)vec.size() && vec[(

)vec.size()-1-three_p] ==

'N'

; ++three_p);

811 if

(five_p > 0 || three_p > 0) {

819 _ASSERT

(m.

().front().Limits().GetLength() > left);

823 _ASSERT

(m.

().back().Limits().GetLength() > right);

827 double

score = m.

Score

();

844

arg_desc->

AddKey

(

"param"

845 "Organism specific parameters"

848

arg_desc->

(

"nognomon"

"Skips ab initio prediction and ab initio extension of partial chains."

);

851

arg_desc->

(

"open"

"Allow partial predictions at the ends of contigs. Used for poorly assembled genomes with lots of unfinished contigs."

);

853

arg_desc->

(

"nonconsens"

"Allows to accept nonconsensus splices starts/stops to complete partial alignmet. If not allowed some partial alignments " 854 "may be rejected if there is no way to complete them."

);

857

arg_desc->

(

"norep"

"DO NOT mask lower case letters"

);

861

arg_desc->