I for one am surprised and wish for others to chip in their 2 cents.
Have u tried using bowtie instead of tophat? Since u are using est for ref. I am curious if bowtie will differ from top hat in this case.

But afaik, one diff betw bowtie and bwa is that bwa allows for indels whilst bowtie doesn't. it might be useful if you can provide more info on the different parameters u used for both ur analysis,

Cheers
Kevin

On 28 Apr 2011 05:40, "Austin Paul" <austinpa@usc.edu> wrote:
> Hi Vasu,
>
> Thanks for your response. I realize tophat and bowtie are different. But
> my question was concerning what seemed to me to be divergent results from
> mapping with tophat and bwa (which I also realize are different). bwa
> seemed to map around 4x the reads that tophat mapped using the same
> reference. Should this not come as a surprise?
>
> Austin
>
> On Wed, Apr 27, 2011 at 2:30 PM, vasu punj <punjv@yahoo.com> wrote:
>
>> Though Tophat calls in Bowtie but they are different mapping tools. Details
>> can be found in mannual of Tophat. For RNA-seq one may be stick with Tophat.
>>
>> Vasu
>>
>> --- On *Wed, 4/27/11, Austin Paul <austinpa@usc.edu>* wrote:
>>
>>
>> From: Austin Paul <austinpa@usc.edu>
>> Subject: [galaxy-user] mapping with tophat vs. bwa
>> To: galaxy-user@lists.bx.psu.edu
>> Date: Wednesday, April 27, 2011, 4:20 PM
>>
>>
>> Hello,
>>
>> I am getting what seems to me to be strange results using two different
>> mapping tools in Galaxy. I am mapping illumina RNA-seq data and with
>> tophat, while setting # alignments to 1, I get around 15-20% reads mapping.
>> And when I use bwa, I am getting around 75% reads mapping. My reference is
>> a collection of ESTs so the strength of tophat being a spliced read mapper
>> is probably not being utilized, but I am surprised by the difference in the
>> number of reads mapping between the two. Any thoughts?
>>
>> Austin
>>
>> -----Inline Attachment Follows-----
>>
>> ___________________________________________________________
>> The Galaxy User list should be used for the discussion of
>> Galaxy analysis and other features on the public server
>> at usegalaxy.org. Please keep all replies on the list by
>> using "reply all" in your mail client. For discussion of
>> local Galaxy instances and the Galaxy source code, please
>> use the Galaxy Development list:
>>
>> http://lists.bx.psu.edu/listinfo/galaxy-dev
>>
>> To manage your subscriptions to this and other Galaxy lists,
>> please use the interface at:
>>
>> http://lists.bx.psu.edu/
>>
>>