Difference between direct vs. indirect contacts

[jychoilab]

Hi Eoghan

Just a clarification question. I was parsing through the alignment stat output at logs/results/merged_contacts/*.concatemers.parquet.log
and saw direct vs. indirect contact numbers. I went back the pore-c biorxiv and saw the following terminologies:

Direct pair: a pair of restriction fragments that are adjacent on the read and is roughly equivalent to what a chimeric Hi-C read detects.

Indirect pair: a pair of restriction fragments that are not adjacent on the read (i.e. those with other restriction fragments between them).

Chimeric Junction: A junction in a concatemer generated by the ligation of two restriction fragment ends that are non-adjacent in the genome. These junctions are used to infer 3D proximity.

Non-Chimeric Junction: A junction in a concatemer generated by the ligation of two restriction fragment ends that are adjacent in the genome. This can be the result of incomplete digestion, the re-ligation of cognate ends of a digestion event or possibly a SNP at the digestion site.

So the Hi-C matrix (cool or hic or whatever you decide to generate) that is generated at the end is only using the direct pairs?

The reason I ask is, it seems if I add concatemer_order up it doesn't seem to match any of the contact counts reported. I have a feeling its because of ignoring the 1-way contact (which is useless info) but wanted to also make sure which contact was being used at the end.

Thanks for the clarification in advance.

Jae

[jychoilab]

Sorry one more question. Whats the distance cutoff for long_range_cis_contacts vs short_range_cis_contacts?