I am a PSU student at the College of Medicine. I have minimal bioinformatics and computer programing background. I was introduced to galaxy during televisied genomics course. I believe I am using galaxy on the cloud. I am trying to display ChIP-CHIP data on UCSC. I have downloaded the gff file (attached; NCBI/GEO # GSM413116) and uploaded it into galaxy (using get data; upload file). The initial publication converted gff file to wig file for upload into UCSC. I could not find this function in Galaxy so I tried converting to bed file instead (convert formats; gff-to-bed). When I do this and upload into UCSC I do not get the peak hight (score difference) at each site (column 6 of gff file; score--which appears to convert to all 0s in bed file), only the locations are annotated but not the score. Is there something I am doing wrong in the conversion to a bed file or alternatively is there a way to convert the intial gff file into a wig file? Any help would be greatly appreciated! Thanks, Lindsay Horvath
Hello Lindsay, There are no tools at Galaxy main or in the Tool Shed (for a local or cloud instance) that will convert a .gff file into a .wig file. A .gff file can be visualized at UCSC directly, without being converted to wig, as long as the format is in spec and the database (mouse in your case) is assigned). Visualization in Galaxy is also an option. Converting from .bed to .gff does not include the score, because BED files have a specific requirement for score values to be between 0-1000. A strict GFF file would also have a score value between 0-1000 (according to the UCSC definition), but others sites state that floating point numbers in general area accepted. The file you are working with has decimal value scores that would be rejected as valid GFF or BED scores at UCSC. So, at a minimum, these values should be scaled and/or rounded off before sending to UCSC for visualization in GFF format. The modified score value could also be swapped into a converted BED file (the one with all "0" for the score column), using tools in the "Text Manipulation" group. Or, you can try transforming the data as the authors in the publication did. Data in .wig format can have associated scores in decimal format. Once you have .wig format, it can be converted to .bigWig, and genome-wide visualization becomes much more efficient between the two servers (Galaxy -> UCSC). Hopefully this helps, Jen Galaxy team On 3/14/12 11:39 AM, Horvath, Lindsay wrote:
I am a PSU student at the College of Medicine. I have minimal bioinformatics and computer programing background. I was introduced to galaxy during televisied genomics course. I believe I am using galaxy on the cloud.
I am trying to display ChIP-CHIP data on UCSC. I have downloaded the gff file (attached; NCBI/GEO # GSM413116) and uploaded it into galaxy (using get data; upload file). The initial publication converted gff file to wig file for upload into UCSC. I could not find this function in Galaxy so I tried converting to bed file instead (convert formats; gff-to-bed). When I do this and upload into UCSC I do not get the peak hight (score difference) at each site (column 6 of gff file; score--which appears to convert to all 0s in bed file), only the locations are annotated but not the score. Is there something I am doing wrong in the conversion to a bed file or alternatively is there a way to convert the intial gff file into a wig file? Any help would be greatly appreciated!
Thanks,
Lindsay Horvath
___________________________________________________________ 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:
participants (2)
-
Horvath, Lindsay
-
Jennifer Jackson