Information about GRCh38 reference genome addition

[poddarharsh15]

Hi @vicsanga

I hope you're doing well. I wanted to ask if there are any updates regarding hg38 support in POSTRE, or should I still need to use liftOver to convert coordinates to hg19 for analysis?

Thanks in advance!
Best,
Harsh

[vicsanga]

Hi @poddarharsh15

I hope you're doing well too!
The new version, which has hg38 as reference genome, will not be ready until the end of the year/beginning of the next one. So, until then, you still need to liftOver the coordinates. Sorry for the inconvenience,

Have a nice day!
Victor

[poddarharsh15]

Hi @vicsanga

I am planning to use POSTRE on a cohort of 100 cardiomyopathy patients. Before running the analysis, I intend to perform a liftover on my VCF files.(GRCh38 to GRCh37)

Could you please clarify whether it is necessary to manually add TAD annotations for these samples, or if the default annotations provided by POSTRE are sufficient (I already have HPO terms for all patients)?

Additionally, do you recommend a specific tool for the liftover process?

Thank you in advance for your guidance.

Best regards,
Harsh Poddar

[vicsanga]

Hi Harsh,
Nice to know! Regarding the different questions:

• You do not need to add manually TAD annotation. For any phenotype available in POSTRE, there are internally some TAD maps associated. Regarding heart abnormalities, the TAD maps currently used are derived of a cardiomyocyte differentiation from: https://doi.org/10.1038/s41588-019-0479-7. You can find more info in the User Guide > Cell Types/Tissues considered per Phenotype section. However, if you are not happy with the available TAD maps for the phenotype of interest you can use/upload your own TAD map (for the first run I would probably give it a try with those available in POSTRE).

• For the liftover process, although there are many good tools available for this, I usually use https://genome.ucsc.edu/cgi-bin/hgLiftOver (there are also some R packages available for this).

Let me know if you have more doubts.

Have a nice day!
Victor

[poddarharsh15]

Hi @vicsanga,
Thank you very much for the clarification. I will proceed with the TAD annotations provided by POSTRE and will reach out if I need any further assistance. (sorry for late response)

Best regards,
Harsh Poddar

[poddarharsh15]

Hi @vicsanga

I’ve tested these data with my files and attached the TSV files generated by POSTRE. I’m confused because I can’t see any TADs information. Could you please check if I might have missed something during the run?
Many thanks!

[PS: If you have time next week I would like to organize a meeting with you for detailed understanding of POSTRE if possible please let me know :) ]

ResultsPerSV.txt
TEST5.tsv

[vicsanga]

Hi @poddarharsh15
sorry for my late reply but I completely missed this answer (not sure why I did not get any notification). I just saw it!

What you are showing me is the aggregated table generated upon a multi SV analysis, for a detailed SV explanation you have to then run a single SV analysis with the SV of interest (and if there is a predicted long-range/enhancer mediated prediction then you will see the TADs, changes in enhancer landscape etc). You can also gather this information in an aggregated manner from additional tables provided.

Please, send me an email to postre.radaiglesiaslab@gmail.com and we can arrange an online meeting, happy to do it!

Best regards,
Victor

[poddarharsh15]

Hi @vicsanga

Thank you for the explanation—that makes perfect sense! I'll send you an email soon (probably next week) so we can arrange a meeting. I'd love to show you some data from my patients and get your insights on the SVs.

No worries at all about the late response—I totally understand. Looking forward to chatting soon!

Best regards,
Harsh