Skip to content

Latest commit

 

History

History
59 lines (38 loc) · 4.02 KB

FAQ.md

File metadata and controls

59 lines (38 loc) · 4.02 KB
Raw

Version issues (1.x vs 2.x)

Juicer and Juicer Tools are under active development, and we have not made official releases for version 2 yet. In general, while we try to maintain backward compatibility, it is not always possible. Some general guidelines:

  • Files built with Juicer Tools 1.x should be readable by all future 2.x jars for most tools (HiCCUPS, Arrowhead, etc).
  • For file writing tools (pre and addnorm) the same jar must be used (i.e. if pre was called from a 1.x jar, addnorm must also be called from a 1.x jar). This is because 2.x jars only write the latest v9 of the .hic file.
  • To upgrade your v8 .hic files to v9, you can either rebuild files using the merged_nodups or use the EMT Tool to build a newer .hic file from an older one.

KR is not available for a particular chromosome/region

The KR algorithm sometimes does not converge for a particular region/chromosome. When the normalization fails for a chromosome at a specific resolution, the .hic file does not contain the normalization vector for that resolution as well as higher resolutions. More recently, we have added a new normalization method, called SCALE to remedy this. You can use addNorm in newer versions of juicer tools to use SCALE instead of KR.

Normalization failing for fragment (FRAG) resolutions or FRAG resolution questions in general

We've actually moved away from fragment resolution maps. Unless you are doing some very specific/specialized development, you probably don't need them. Going forward, Juicer 2 will no longer build fragment resolutions. Instead, you can build higher basepair (BP) resolution maps. Remove the -f flag for pre or add the -F flag for addnorm to ignore these FRAG resolutions.

Juicer pre is taking a very long time

If you are assembling a genome, don't try to build a hic map with thousands of contigs. Checkout the very detailed walkthrough on the DNAZoo Website.

How can I access published Hi-C data?

You can use Juicebox and Juicer command line tools directly on URLs. Check out our data page for those links. We recommend this method as opposed to downloading full .hic files, but if there's an experiment you plan to explore a lot and/or you have a bad internet connection, you can download locally.

If you would like to download a local version of a .hic file, check out the GEO repository for the experiment. We have also put some maps up in a Box mirror

You can also parse the default properties file at http://hicfiles.tc4ga.com/juicebox.properties to grab the web address of the file you're interested in.

How do I create my own .hic files from my own data?

Check out the Juicer command line tools pre command

What's the format of the restriction site file?

See these directions

Error: the chromosome combination appears in multiple blocks

Please see this thread

Juicer (pre) is slow and/or takes a lot of memory

Try running without a fragment map. For the Juicer pipeline, use the flag -x to exclude fragment maps. At the Pre stage, simply don't send in the restriction site file.

If you are only interested in a few resolutions, you can call Pre with the -r flag. E.g., to create a file with only 25Kb and 10Kb binned maps, do -r 5000,10000

I have already binned and/or normalized data, how do I create a .hic file?

Use Pre with the nine column "score" format. See https://github.com/theaidenlab/juicer/wiki/Pre#short-with-score-format

My question wasn't answered here

Please search the 3D genomics forum and ask your question there