recreate metadata.xml

[tolot27]

Hi!

Is there any way to reconstruct the metadata.xml?

Many attributes are stored inside the bax.h5, i. e. at ScanData/RunInfo, but not the adapter sequence.

Background: I got some SMRTCell data (bas.h5 and bax.h5) but no metadata.xml and currently there is no way to get the metadata.xml. That's way I want to reconstruct it, at least partially so that I can import the SMRTCells into SMRTAnalysis.

--
Kind regards,
Mathias

[GenoMax]

Dr. Hall from PacBio participates in the forum and may have a more definitive answer for you later today.

Can't you ask for the metadata.xml (since you have got the .h5 files) from the provider? The format for the metadata.xml file is described here: http://files.pacb.com/software/instr...ut%20Guide.pdf. Looking at an example file I am not sure if you can reconstruct one from scratch without knowing a lot of additional information about the instrument/reagent lots/run (not sure how much of the information is absolutely required).

[rhall]

Generating a valid metadata.xm from scratch is very difficult and not recommended. If it really is not possible to get the original file I would suggest taking one from an experiment that uses the same chemistry and making alterations to that file. A good place to start would be PacBio's list of datasets:

https://github.com/PacificBiosciences/DevNet/wiki/Datasets

The file would need to be renamed to match the bax.h5 files, and minimally the RunId and Sample Name would have to be corrected.
Unfortunately multiple algorithms use the chemistry information in this file that is defined by the lot barcodes, so it is very important to use the correct information.