Re: Questions concerning seed selection in Bartender paper (Bioinformatics 2018)

[赵路]

Dear Pina Fanny Ida,

Thanks for your interests in our Bartender tool. The "Sequence" in Figure S2 refers to the extracted barcode sequence not the original raw reads. And "reads" in figure S2 also refers the extracted barcode sequence not the raw reads. Probably "Extracted Barcodes" might be a better name for both places.

To be clear, the clustering algorithm always starts with the barcodes extracted from the raw reads. One of the setup steps of the clustering algorithm is to compute and select seed positions using entropy values of all positions in the barcode sequence. And the seed positions is used to distribute clusters to different bins. And the other setup step is to form the initial cluster-list using the unique extracted barcodes with their frequencies as the initial cluster sizes.  These two steps are independent and could be executed in arbitrary order.

Hopefully I answer your question clearly. If you have further questions about Bartender tool and its algorithm, please post your question to this google group in case other users also have similar/same questions. https://groups.google.com/forum/#!forum/bartenderrandombarcode

Cheers,

Lu

On Fri, Mar 15, 2019 at 3:09 PM Sasha F. Levy <sfl...@stanford.edu> wrote:

Hi Pina,

Sorry for the delay. I have changed emails and rarely check this one. I’m CCing the first authors on this who can better answer you question.

Best,

Sasha

On Mar 6, 2019, at 3:51 AM, Krell, Pina Fanny Ida <pina....@uni-bielefeld.de> wrote:

Dear Mr. Levy,

I am currently trying to integrate a citation to your tool Bartender into a review for my 

PhD, but am experiencing some trouble with the details concerning the algorithmic 

description in the Bartender paper.

The supplement to the paper describes the algorithm to work with seeds extracted from 

the read sequences.

Contradicting the pseudocode for the algorithm (figure S2) describes the initial cluster-list 

based on the barcode sequence frequency.

As it is a relevant difference on which part of the sequence the reads seeds will be selected

and reads will be clustered, i was wondering maybe you could help me clarify the matter.

Thank you in advance for your time,

Kind regards,

Pina Krell

--

Sincerely,

 

Lu