package Bio::EnsEMBL::Funcgen::PipeConfig::ProbeMapping::AlignProbes_conf; use strict; use warnings; use Bio::EnsEMBL::Hive::PipeConfig::HiveGeneric_conf; # Allow this particular config to use conditional dataflow and INPUT_PLUS use base ('Bio::EnsEMBL::Funcgen::PipeConfig::ProbeMapping::Base'); my $max_allowed_hits_per_probe = 100; sub pipeline_analyses { my $self = shift; return [ { -logic_name => 'start_align_probes', -module => 'Bio::EnsEMBL::Hive::RunnableDB::Dummy', -flow_into => { MAIN => 'truncate_probe_feature_tables', }, }, { -logic_name => 'truncate_probe_feature_tables', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SqlCmd', -parameters => { sql => [ "truncate probe_feature;", "truncate probe_transcript;", "truncate probe_set_transcript;", "truncate probe_feature_transcript;", "truncate unmapped_object;", "truncate unmapped_reason;", ], db_conn => 'funcgen:#species#', }, -flow_into => { MAIN => 'switch_probe_feature_table_to_innodb', }, }, { -logic_name => 'switch_probe_feature_table_to_innodb', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SqlCmd', -parameters => { sql => [ "ALTER TABLE probe_feature ENGINE=InnoDB;", ], db_conn => 'funcgen:#species#', }, -flow_into => { 'MAIN->A' => 'split_into_probe_chunks', 'A->MAIN' => 'done_processing_probe_chunks', }, }, { -logic_name => 'split_into_probe_chunks', -module => 'Bio::EnsEMBL::Hive::RunnableDB::FastaFactory', -parameters => { inputfile => '#unmapped_sequences_file#', output_dir => '#tempdir#/#species#/probe_chunks', max_chunk_length => 100000, output_prefix => 'probe_chunk_', output_suffix => '.fasta', hash_directories => 1, }, -flow_into => { 2 => [ { 'multiply_probe_align_jobs' => INPUT_PLUS }, ], # '2->A' => [ { 'multiply_probe_align_jobs' => INPUT_PLUS }, ], # 'A->2' => [ { 'done_processing_probe_chunks' => undef }, ] }, }, { -logic_name => 'multiply_probe_align_jobs', -module => 'Bio::EnsEMBL::Funcgen::RunnableDB::ProbeMapping::AddJobParameter', -parameters => { add => { species => '#species#', } }, -flow_into => { 2 => [ 'start_align_probes_genomic', 'start_align_probes_transcript' ], }, }, { -logic_name => 'start_align_probes_genomic', -module => 'Bio::EnsEMBL::Funcgen::RunnableDB::ProbeMapping::AddJobParameter', -parameters => { add => { type => 'genomic', } }, -flow_into => { 2 => 'probe_align_genomic', }, }, { -logic_name => 'start_align_probes_transcript', -module => 'Bio::EnsEMBL::Funcgen::RunnableDB::ProbeMapping::AddJobParameter', -parameters => { add => { type => 'transcript', } }, -flow_into => { 2 => 'probe_align_transcript', }, }, { -logic_name => 'probe_align_genomic', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SystemCmd', -rc_name => '8Gb_job', # -analysis_capacity => 1, -batch_size => 10, -parameters => { cmd => ' exonerate \ --querytype dna \ --targettype dna \ --query #chunk_name# \ --target #toplevel_sequences_file# \ --showsugar false --showvulgar false --showalignment false --ryo "RESULT: %S %pi %ql %tl %em scores:{%Ps:}\\n" \ > #chunk_name#_genomic.exonerate.txt ' }, -flow_into => { MAIN => 'start_store_probe_feature_chunk', '-1' => 'probe_align_genomic_himem', }, }, { -logic_name => 'probe_align_genomic_himem', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SystemCmd', -rc_name => '64Gb_job', -batch_size => 10, -parameters => { cmd => ' exonerate \ --querytype dna \ --targettype dna \ --query #chunk_name# \ --target #toplevel_sequences_file# \ --showsugar false --showvulgar false --showalignment false --ryo "RESULT: %S %pi %ql %tl %em scores:{%Ps:}\\n" \ > #chunk_name#_genomic.exonerate.txt ' }, -flow_into => { MAIN => 'start_store_probe_feature_chunk', }, }, { -logic_name => 'probe_align_transcript', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SystemCmd', -batch_size => 10, -rc_name => '8Gb_job', # -analysis_capacity => 1, -parameters => { cmd => ' exonerate \ --querytype dna \ --targettype dna \ --query #chunk_name# \ --target #gene_sequences_file# \ --showsugar false --showvulgar false --showalignment false --ryo "RESULT: %S %pi %ql %tl %em scores:{%Ps:}\\n" \ > #chunk_name#_transcript.exonerate.txt ' }, -flow_into => { MAIN => 'start_store_probe_feature_chunk', '-1' => 'probe_align_transcript_himem', }, }, { -logic_name => 'probe_align_transcript_himem', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SystemCmd', -batch_size => 10, -rc_name => '64Gb_job', -parameters => { cmd => ' exonerate \ --querytype dna \ --targettype dna \ --query #chunk_name# \ --target #gene_sequences_file# \ --showsugar false --showvulgar false --showalignment false --ryo "RESULT: %S %pi %ql %tl %em scores:{%Ps:}\\n" \ > #chunk_name#_transcript.exonerate.txt ' }, -flow_into => { MAIN => 'start_store_probe_feature_chunk', }, }, { -logic_name => 'start_store_probe_feature_chunk', -module => 'Bio::EnsEMBL::Hive::RunnableDB::Dummy', }, { -logic_name => 'done_processing_probe_chunks', -module => 'Bio::EnsEMBL::Hive::RunnableDB::Dummy', -flow_into => { MAIN => 'delete_redundant_probe_features', }, }, { -logic_name => 'delete_redundant_probe_features', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SystemCmd', -parameters => { cmd => ' delete_redundant_probe_features.pl \ --registry #reg_conf# \ --species #species# ', }, -flow_into => { MAIN => 'delete_probe_features_on_transcripts_from_methylation_arrays', }, }, { -logic_name => 'delete_probe_features_on_transcripts_from_methylation_arrays', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SqlCmd', -parameters => { sql => [ 'delete from probe_feature using probe_feature join probe using (probe_id) join array_chip using (array_chip_id) join array using (array_id) where array.format="METHYLATION" and probe_feature.source="transcript"', ], db_conn => 'funcgen:#species#', }, -flow_into => { MAIN => 'delete_probe_features_from_known_promiscuous_probes', }, }, { -logic_name => 'delete_probe_features_from_known_promiscuous_probes', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SqlCmd', -parameters => { sql => [ 'drop table if exists temp_promiscuous_probes_with_features;', " create table temp_promiscuous_probes_with_features as select distinct unmapped_object.ensembl_id as probe_id from unmapped_object join unmapped_reason on ( unmapped_object.unmapped_reason_id=unmapped_reason.unmapped_reason_id and summary_description = 'Promiscuous probe' ) join probe_feature on (probe_id = ensembl_id and ensembl_object_type = 'Probe'); ", 'delete from probe_feature using probe_feature join temp_promiscuous_probes_with_features using(probe_id);', 'drop table if exists temp_promiscuous_probes_with_features;', ], db_conn => 'funcgen:#species#', }, -flow_into => { MAIN => 'delete_probe_features_from_promiscuous_probes', }, }, # This will typically pick up on probes that have scored less than the # $max_allowed_hits_per_probe probe features for genomic hits and less # than $max_allowed_hits_per_probe for transcript hits. But when both # numbers are combined, they exceed $max_allowed_hits_per_probe and # the probe is classified as promiscuous after all. # # This analysis will find probes of this kind, remove their probe # features and create an unmapped object summarising the number of # probe features from transcript and genomic matches. # { -logic_name => 'delete_probe_features_from_promiscuous_probes', -module => 'Bio::EnsEMBL::Hive::RunnableDB::SystemCmd', -parameters => { cmd => ' delete_promiscuous_probe_features.pl \ --species #species# \ --registry #reg_conf# \ --analysis_logic_name ProbeAlign_genomic \ --max_allowed_hits_per_probe ' . $max_allowed_hits_per_probe, }, }, ]; } 1;