# Copyright (c) 2013, The Developers # All rights reserved. # Default values for all options that take an argument can be defined here. Any of these values can be overriden # by supplying the corresponding option on the command line. File paths can be absolute or relative to the current # working directory. [Project] # A short name to identify the project, often an abbreviated genome name and a version number, e.g. Aspni3v2. Required. ProjectName= (None) # Path to the root directory for project intermediate and result files; defaults to SnowyOwl/Projects/ ProjectDir= (None) # Path to the genome sequence file in FASTA format. Required. Genome= (None) # Path to a genome sequence file with locations where no gene predictions are wanted (e.g. ribosomal DNA, transposons, # known pseudogenes) masked with N. Optional. MaskedGenome= (None) # Path to a file containing RNA-Seq reads in FASTQ or FASTA format for providing gene-prediction hints. Required. Reads= (None) # Path to a directory containing the results of mapping RNA-Seq reads to the genome. It must include tabix-indexed files # containing splice junctions, named classified_juncs.gz, and coverage depth, named tuque.coverage.wig.gz. Suitable files # can be prepared from mapped reads in BAM format with the included script SnowyOwl/bin/scripts/BAM_to_juncs_and_coverage.sh. # Required. MappedReads= (None) # Path to a file in FASTA format containing predicted transcript sequences, usually contigs assembled from RNA-Seq reads. # Required Transcripts= (None) # A short label used as a prefix for the names of predicted genes; it defaults to the project name. Optional label= (None) [Execution] # Number of processes to use; if 0, it is set to 1 less than the total number of processors on the system. Optional. np=0 # Blast search type. The choices are: # Software - Blast+ executed in parallel on your local system, # DClocal - Hardware-accelerated tera-blast search on a DeCypher TimeLogic board on your local system, # DCremote - Hardware-accelerated tera-blast search on a DeCypher TimeLogic board on a remote server BlastType=Software [Blast] ## For Blast searching without hardware acceleration ## # Protein database for blastx searching with potential transcript sequences. # Prepared from downloaded FASTA file(s) with makeblastdb command blastx_db=/export/geno_new/SO_subProg_Database/Database/uniprot_sprot_2010_11 # Protein database for homology scoring of predicted protein sequences. # Prepared from downloaded FASTA file(s) with makeblastdb command blastp_db=/export/geno_new/SO_subProg_Database/Database/Fungal_RefSeq [Tera-blast] ## For hardware-accelerated Blast searching with DeCypher TimeLogic boards ## # DeCypher blastx target database name dc_blastx_targ=uniprot_sprot_2010_11 # DeCypher blastp target database name dc_blastp_targ=Fungal_RefSeq # DeCypher machine (board) name as known to the server dc_mach=coedc06 ## DClocal # Paths to local tera-blast template files dc_local_blastx_template=/export/data_new/programs/decypher/templates/tera-blastx dc_local_blastp_template=/export/data_new/programs/decypher/templates/tera-blastp.refseq # DeCypher program path on local server dc_local_path=/export/data/linux/decypher/bin ## DCremote # Server on which DeCypher boards are remotely accessed dc_server=coe01.ucalgary.ca # User name on the server dc_user= # DeCypher program path on remote server dc_remote_path=/export/data_new/programs/decypher/bin # Paths to remote tera-blast template files dc_remote_blastx_template=blast/tera-blastx dc_remote_blastp_template=blast/tera-blastp.refseq MAX_TIMEOUTS=5 ; Maximum number of timeouts during file transfer [Exonerate] # Smallest allowed intron size for alignment of transcripts to genome. imin=50 # Largest allowed intron size for alignment of transcripts to genome. imax=2000 [Scoring] # Models with lower mean coverage are assigned "Low expression" status DEPTH_THRESHOLD=10 # Score initially assigned to all models DEFAULT_SCORE=1.0 # Maximum number of homologs used for scoring MAX_HOMOLOGS=3 # Shorter predicted proteins from monoexonic models with no homologs in database are penalized MIN_PROTEIN_LENGTH=150 # Transcripts containing an exon with a median coverage depth that is a smaller fraction of the transcript median depth # are assigned a status of Heterogeneous coverage depth MIN_MEDIAN_RATIO=0.15 #Transcripts containing an exon with a median coverage depth that is a larger multiple of the transcript median depth # are assigned a status of Heterogeneous coverage depth MAX_MEDIAN_RATIO=3 # Transcripts containing a longer run of coverage depth less than coverage_tolerance * transcript mean depth are assigned # a status of Heterogeneous coverage depth MAX_LOW_RUN=10 # Fraction of transcript mean depth used to define a low run COVERAGE_TOLERANCE=0.05 # Splices with a higher readthrough ratio are ignored during scoring MAX_INTRON_READTHROUGH=0.8 # Predicted introns shorter than this are rejected MIN_INTRON_LENGTH=10 # Predicted introns longer than this are rejected MAX_INTRON_LENGTH=2000 # Transcripts with more Ns in their sequences are assigned a status of Cannot be translated into protein MAX_UNKNOWN_BASES=8 # Minimum score for a model to be considered correct CORRECT_SCORE_THRESHOLD=0.5 [Merging] # A small positive constant added to all scores to avoid merging problems MERGE_PSEUDOSCORE=0.0001 [Splice_hinted_Augustus] # Minimum coverage depth inside a block of contiguous coverage MIN_COVER=3 # Transcripts shorter than this are not reported MIN_TRANSCRIPT_LENGTH=100 # Exons shorter than this are not subdivided because of jumps in coverage depth MIN_EXON_LENGTH=50 # Splices with read-through ratios higher than this are not used to create introns between read islands MAX_SPLICE_READTHROUGH=0.6 # Any coverage gap longer than this causes a new locus to be started MAX_INTRAGENE_GAP=500 # Regions for Augustus gene prediction are extended by this much at each end REGION_PAD=0 [Config] # Path to the CONFIG file config_file=