tophat building bowtie index from genome.fa

Question

[jiangquanlong]

I already build the index by bowtie. why when tophat run, it will build bowtie index from genome.fa again?? (the tophat can run normally. I just want to why and if there is someways to save the time.)

[

Mon Jul  4 16:41:37 2016] Beginning TopHat run (v1.4.1)
-----------------------------------------------
[Mon Jul  4 16:41:37 2016] Preparing output location /home/jiangquanlong/test_data/out2/
[Mon Jul  4 16:41:37 2016] Checking for Bowtie index files
[Mon Jul  4 16:41:37 2016] Checking for reference FASTA file
[Mon Jul  4 16:41:37 2016] Checking for Bowtie
        Bowtie version:                  1.1.1.0
[Mon Jul  4 16:41:37 2016] Checking for Samtools
        Samtools Version: 0.1.16
[Mon Jul  4 16:41:37 2016] Generating SAM header for /home/jiangquanlong/jql_han/0.Genome/WBcel235/WBcel235
        format:          fastq
        quality scale:   phred33 (default)
[Mon Jul  4 16:41:39 2016] Reading known junctions from GTF file
[Mon Jul  4 16:41:54 2016] Preparing reads
        left reads: min. length=100, count=3998727
[Mon Jul  4 16:48:11 2016] Creating transcriptome data files..
****[Mon Jul  4 16:48:22 2016] Building Bowtie index from WBcel235.fa****
[Mon Jul  4 16:54:25 2016] Mapping left_kept_reads against transcriptome WBcel235 with Bowtie 
[Mon Jul  4 17:01:04 2016] Converting left_kept_reads.m2g to genomic coordinates (map2gtf)
[Mon Jul  4 17:05:51 2016] Resuming TopHat pipeline with unmapped reads
[Mon Jul  4 17:05:51 2016] Mapping left_kept_reads.m2g_um against WBcel235 with Bowtie