-G and -j option

[HaiHuaLiu]

Hi,
Here, I have some questions. Firstly, I want to provide -G option a gtf file which includes four annotation files(Ensembl, RefSeq gene annotations, spliced ESTs and mRNAs obtained from the UCSC genome browser). So, the gtf file contains duplicate records without sorting. Does it matter? Or I need to sort the file first and remove the duplicates.

Secondly, does TopHat just use the exon information from gtf file?

Thirdly, I would like to know more details about the differences between -G and -j option. I have read the manual, still I am confused about the usages of these two options and their functions. I need to run TopHat two-pass mode, for example,

On the first round, I use -G option to supply the reference annotation gtf file, the commander shows as below:

tophat2 -p 8 -G hg19_annotation.gtf hg19 T1_read1.fq,T2_read1.fq T1_read2.fq,T2_read2.fq

On the second round, I use -j option to supply the junctions file produced by bed_to_juncs in the first round, the commander shows as below:

topgat2 -p 8 -j new_list.juncs hg19 T1_read1.fq T1_read2.fq

I wonder the above workflow works and the -j option on the second round functions in the same way as -G option on the first round.

Look forward to your reply.

Best regards.

Haihua