Galaxy | Tool Preview

DeFuse (version 0.6.2.2)
The left and right reads pairs must be in the same order, and not have any unpaired reads. (FASTQ interlacer will pair reads and remove the unpaired. FASTQ de-interlacer will separate the result into left and right reads.)
In the same order as the left reads
Value to put in the results library_name column
if your genome of interest is not listed - contact Galaxy team
The defuse output working directory can be helpful for determining errors that may have occurred during the run, but they require considerable diskspace, and should be deleted and purged when no longer needed.

DeFuse

DeFuse is a software package for gene fusion discovery using RNA-Seq data. The software uses clusters of discordant paired end alignments to inform a split read alignment analysis for finding fusion boundaries. The software also employs a number of heuristic filters in an attempt to reduce the number of false positives and produces a fully annotated output for each predicted fusion.

Journal reference: http://www.ploscompbiol.org/article/info%3Adoi%2F10.1371%2Fjournal.pcbi.1001138

Inputs

DeFuse requires 2 fastq files for paried reads, one with the left mate of the paired reads, and a second fastq with the the right mate of the paired reads (with reads in the same order as in the first fastq dataset).

If your fastq files have reads in different orders or include unpaired reads, you can preprocess them with FASTQ interlacer to create a single interlaced fastq dataset with only the paired reads and input that to FASTQ de-interlacer to separate the reads into a left fastq and right fastq.

DeFuse uses a Reference Dataset to search for gene fusions. The Reference Dataset is generated from the following sources in DeFuse_Version_0.4:
  • genome_fasta from Ensembl
  • gene_models from Ensembl
  • repeats_filename from UCSC RepeatMasker rmsk.txt
  • est_fasta from UCSC
  • est_alignments from UCSC intronEst.txt
  • unigene_fasta from NCBI

Outputs

The galaxy history will contain 5 outputs: the config.txt file that provides DeFuse with its parameters, the defuse.log which details what DeFuse has done and can be useful in determining any errors, and the 3 results files that defuse generates.

DeFuse generates 3 results files: results.txt, results.filtered.txt, and results.classify.txt. All three files have the same format, though results.classify.txt has a probability column from the application of the classifier to results.txt, and results.filtered.txt has been filtered according to the threshold probability as set in config.txt.

The file format is tab delimited with one prediction per line, and the following fields per prediction (not necessarily in this order):

  • Identification
    • cluster_id : random identifier assigned to each prediction
    • library_name : library name given on the command line of defuse
    • gene1 : ensembl id of gene 1
    • gene2 : ensembl id of gene 2
    • gene_name1 : name of gene 1
    • gene_name2 : name of gene 2
  • Evidence
    • break_predict : breakpoint prediction method, denovo or splitr, that is considered most reliable
    • concordant_ratio : proportion of spanning reads considered concordant by blat
    • denovo_min_count : minimum kmer count across denovo assembled sequence
    • denovo_sequence : fusion sequence predicted by debruijn based denovo sequence assembly
    • denovo_span_pvalue : p-value, lower values are evidence the prediction is a false positive
    • gene_align_strand1 : alignment strand for spanning read alignments to gene 1
    • gene_align_strand2 : alignment strand for spanning read alignments to gene 2
    • min_map_count : minimum of the number of genomic mappings for each spanning read
    • max_map_count : maximum of the number of genomic mappings for each spanning read
    • mean_map_count : average of the number of genomic mappings for each spanning read
    • num_multi_map : number of spanning reads that map to more than one genomic location
    • span_count : number of spanning reads supporting the fusion
    • span_coverage1 : coverage of spanning reads aligned to gene 1 as a proportion of expected coverage
    • span_coverage2 : coverage of spanning reads aligned to gene 2 as a proportion of expected coverage
    • span_coverage_min : minimum of span_coverage1 and span_coverage2
    • span_coverage_max : maximum of span_coverage1 and span_coverage2
    • splitr_count : number of split reads supporting the prediction
    • splitr_min_pvalue : p-value, lower values are evidence the prediction is a false positive
    • splitr_pos_pvalue : p-value, lower values are evidence the prediction is a false positive
    • splitr_sequence : fusion sequence predicted by split reads
    • splitr_span_pvalue : p-value, lower values are evidence the prediction is a false positive
  • Annotation
    • adjacent : fusion between adjacent genes
    • altsplice : fusion likely the product of alternative splicing between adjacent genes
    • break_adj_entropy1 : di-nucleotide entropy of the 40 nucleotides adjacent to the fusion splice in gene 1
    • break_adj_entropy2 : di-nucleotide entropy of the 40 nucleotides adjacent to the fusion splice in gene 2
    • break_adj_entropy_min : minimum of break_adj_entropy1 and break_adj_entropy2
    • breakpoint_homology : number of nucleotides at the fusion splice that align equally well to gene 1 or gene 2
    • breakseqs_estislands_percident : maximum percent identity of fusion sequence alignments to est islands
    • cdna_breakseqs_percident : maximum percent identity of fusion sequence alignments to cdna
    • deletion : fusion produced by a genomic deletion
    • est_breakseqs_percident : maximum percent identity of fusion sequence alignments to est
    • eversion : fusion produced by a genomic eversion
    • exonboundaries : fusion splice at exon boundaries
    • expression1 : expression of gene 1 as number of concordant pairs aligned to exons
    • expression2 : expression of gene 2 as number of concordant pairs aligned to exons
    • gene_chromosome1 : chromosome of gene 1
    • gene_chromosome2 : chromosome of gene 2
    • gene_end1 : end position for gene 1
    • gene_end2 : end position for gene 2
    • gene_location1 : location of breakpoint in gene 1
    • gene_location2 : location of breakpoint in gene 2
    • gene_start1 : start of gene 1
    • gene_start2 : start of gene 2
    • gene_strand1 : strand of gene 1
    • gene_strand2 : strand of gene 2
    • genome_breakseqs_percident : maximum percent identity of fusion sequence alignments to genome
    • genomic_break_pos1 : genomic position in gene 1 of fusion splice / breakpoint
    • genomic_break_pos2 : genomic position in gene 2 of fusion splice / breakpoint
    • genomic_strand1 : genomic strand in gene 1 of fusion splice / breakpoint, retained sequence upstream on this strand, breakpoint is downstream
    • genomic_strand2 : genomic strand in gene 2 of fusion splice / breakpoint, retained sequence upstream on this strand, breakpoint is downstream
    • interchromosomal : fusion produced by an interchromosomal translocation
    • interrupted_index1 : ratio of coverage before and after the fusion splice / breakpoint in gene 1
    • interrupted_index2 : ratio of coverage before and after the fusion splice / breakpoint in gene 2
    • inversion : fusion produced by genomic inversion
    • orf : fusion combines genes in a way that preserves a reading frame
    • probability : probability produced by classification using adaboost and example positives/negatives (only given in results.classified.txt)
    • read_through : fusion involving adjacent potentially resulting from co-transcription rather than genome rearrangement
    • repeat_proportion1 : proportion of the spanning reads in gene 1 that span a repeat region
    • repeat_proportion2 : proportion of the spanning reads in gene 2 that span a repeat region
    • max_repeat_proportion : max of repeat_proportion1 and repeat_proportion2
    • splice_score : number of nucleotides similar to GTAG at fusion splice
    • num_splice_variants : number of potential splice variants for this gene pair
    • splicing_index1 : number of concordant pairs in gene 1 spanning the fusion splice / breakpoint, divided by number of spanning reads supporting the fusion with gene 2
    • splicing_index2 : number of concordant pairs in gene 2 spanning the fusion splice / breakpoint, divided by number of spanning reads supporting the fusion with gene 1

Example

results.tsv:

cluster_id    splitr_sequence splitr_count    splitr_span_pvalue      splitr_pos_pvalue       splitr_min_pvalue       adjacent        altsplice       break_adj_entropy1      break_adj_entropy2      break_adj_entropy_min   break_predict   breakpoint_homology     breakseqs_estislands_percident  cdna_breakseqs_percident        concordant_ratio        deletion        est_breakseqs_percident eversion        exonboundaries  expression1     expression2     gene1   gene2   gene_align_strand1      gene_align_strand2      gene_chromosome1        gene_chromosome2        gene_end1       gene_end2       gene_location1  gene_location2  gene_name1      gene_name2      gene_start1     gene_start2     gene_strand1    gene_strand2    genome_breakseqs_percident      genomic_break_pos1      genomic_break_pos2      genomic_strand1 genomic_strand2 interchromosomal        interrupted_index1      interrupted_index2      inversion       library_name    max_map_count   max_repeat_proportion   mean_map_count  min_map_count   num_multi_map   num_splice_variants     orf     read_through    repeat_proportion1      repeat_proportion2      span_count      span_coverage1  span_coverage2  span_coverage_max       span_coverage_min       splice_score    splicing_index1 splicing_index2
1169  GCTTACTGTATGCCAGGCCCCAGAGGGGCAACCACCCTCTAAAGAGAGCGGCTCCTGCCTCCCAGAAAGCTCACAGACTGTGGGAGGGAAACAGGCAGCAGGTGAAGATGCCAAATGCCAGGATATCTGCCCTGTCCTTGCTTGATGCAGCTGCTGGCTCCCACGTTCTCCCCAGAATCCCCTCACACTCCTGCTGTTTTCTCTGCAGGTTGGCAGAGCCCCATGAGGGCAGGGCAGCCACTTTGTTCTTGGGCGGCAAACCTCCCTGGGCGGCACGGAAACCACGGTGAGAAGGGGGCAGGTCGGGCACGTGCAGGGACCACGCTGCAGG|TGTACCCAACAGCTCCGAAGAGACAGCGACCATCGAGAACGGGCCATGATGACGATGGCGGTTTTGTCGAAAAGAAAAGGGGGAAATGTGGGGAAAAGCAAGAGAGATCAGATTGTTACTGTGTCTGTGTAGAAAGAAGTAGACATGGGAGACTCCATTTTGTTCTGTACTAAGAAAAATTCTTCTGCCTTGAGATTCGGTGACCCCACCCCCAACCCCGTGCTCTCTGAAACATGTGCTGTGTCCACTCAGGGTTGAATGGATTAAGGGCGGTGCGAGACGTGCTTT    2       0.000436307890680442    0.110748295953850       0.0880671602973091      N       Y       3.19872427442695        3.48337348351473        3.19872427442695        splitr  0       0       0       0       Y       0       N       N       0       0       ENSG00000105549 ENSG00000213753 +       -       19      19      376013  59111168        intron  upstream        THEG    AC016629.2      361750  59084870        -       +       0       375099  386594  +       -       N       8.34107429512245        -       N       output_dir      82      0.677852348993289       40.6666666666667        1       11      1       N       N       0.361271676300578       0.677852348993289       12      0.758602776578432       0.569678713445872       0.758602776578432       0.569678713445872       2       0.416666666666667       -
3596  TGGGGGTTGAGGCTTCTGTTCCCAGGTTCCATGACCTCAGAGGTGGCTGGTGAGGTTATGACCTTTGCCCTCCAGCCCTGGCTTAAAACCTCAGCCCTAGGACCTGGTTAAAGGAAGGGGAGATGGAGCTTTGCCCCGACCCCCCCCCGTTCCCCTCACCTGTCAGCCCGAGCTGGGCCAGGGCCCCTAGGTGGGGAACTGGGCCGGGGGGCGGGCACAAGCGGAGGTGGTGCCCCCAAAAGGGCTCCCGGTGGGGTCTTGCTGAGAAGGTGAGGGGTTCCCGGGGCCGCAGCAGGTGGTGGTGGAGGAGCCAAGCGGCTGTAGAGCAAGGGGTGAGCAGGTTCCAGACCGTAGAGGCGGGCAGCGGCCACGGCCCCGGGTCCAGTTAGCTCCTCACCCGCCTCATAGAAGCGGGGTGGCCTTGCCAGGCGTGGGGGTGCTGCC|TTCCTTGGATGTGGTAGCCGTTTCTCAGGCTCCCTCTCCGGAATCGAACCCTGATTCCCCGTCACCCGTGGTCACCATGGTAGGCACGGCGACTACCATCGAAAGTTGATAGGGCAGACGTTCGAATGGGTCGTCGCCGCCACGGGGGGCGTGCGATCAGCCCGAGGTTATCTAGAGTCACCAAAGCCGCCGGCGCCCGCCCCCCGGCCGGGGCCGGAGAGGGGCTGACCGGGTTGGTTTTGATCTGATAAATGCACGCATCCCCCCCGCGAAGGGGGTCAGCGCCCGTCGGCATGTATTAGCTCTAGAATTACCACAGTTATCCAAGTAGGAGAGGAGCGAGCGACCAAAGGAACCATAACTGATTTAATGAGCCATTCGCAGTTTCACTGTACCGGCCGTGCGTACTTAGACATGCATGGCTTAATCTTTGAGACAAGCATATGCTACTGGCAGG  250     7.00711162298275e-72    0.00912124762512338     0.00684237452309549     N       N       3.31745197152461        3.47233119514066        3.31745197152461        splitr  7       0.0157657657657656      0       0       N       0.0135135135135136      N       N       0       0       ENSG00000156860 ENSG00000212932 -       +       16      21      30682131        48111157        coding  upstream        FBRS    RPL23AP4        30670289        48110676        +       +       0.0157657657657656      30680678        9827473 -       +       Y       -       -       N       output_dir      2       1       1.11111111111111        1       1       1       N       N       0       1       9       0.325530693397641       0.296465452915709       0.325530693397641       0.296465452915709       2       -       -