`readHic` no data found

[nekramer]

Moving this comment to a separate issue:
Hi, Nicole,

I am using this tool to read a hic file as input "readHic("C:/projects/hic/hic/public_data/hela_s3/Hela_s3_microC_4DNFIVA2IV88.hic", chrom="1", chromstart = 72400000, chromend = 72405000 )", however, it reports this error: Warning: No data found in region. Suggestions: check that chromosome names match genome assembly; check region; check available resolutions with strawr::readHicBpResolutions().. Am I reading the input in a wrong way? Thanks!

Originally posted by @jiangshan529 in #66 (comment)

[nekramer]

Hi @jiangshan529,

There is nothing wrong with the syntax of your current call to the function, but it is possible that you may have to slightly change some parameters based on your unique .hic file or check out some other aspects of your file/genomic region. Here are a couple things you can check:

1. What is the output of using strawr::readHicChroms() with your file? Do these chromosome names match the names of the assembly you are inputing? This may indicate that you need to change chrom = "1" to something like chrom = "chr1".
2. What is the output of using strawr::readHicNormTypes()? Included normalizations are another feature that can differ between .hic files based on processing. The default norm value of readHic is "KR", but it is possible this may not be compatible with your file. If you don't have any normalizations, you can try "NONE".
3. Are you reading the data in at an appropriate resolution, and is that resolution available to you (by checking with strawr::readHicBpResolutions())? I notice that the the region you are reading in is 5000 bp, which is actually quite small and will require a small resolution. Are resolutions smaller than 5000 bp processed in your file?
4. Is your file downloaded and available to your R session? I have had issues with things like Dropbox in the past when the file is only available in the cloud and the function cannot read it.

If none of these troubleshooting tips get you any data, it is also a valid possibility that there is no data in this region. As I mentioned above, this is a small region, and even if you read in the data with something like a 5000 bp resolution, you would only get one pixel of information. If this pixel happens to have no data, then you truly have no data for that region.

I hope these tips help!

Best,
Nicole

[jiangshan529]

Hi @jiangshan529,

There is nothing wrong with the syntax of your current call to the function, but it is possible that you may have to slightly change some parameters based on your unique .hic file or check out some other aspects of your file/genomic region. Here are a couple things you can check:

1. What is the output of using strawr::readHicChroms() with your file? Do these chromosome names match the names of the assembly you are inputing? This may indicate that you need to change chrom = "1" to something like chrom = "chr1".
2. What is the output of using strawr::readHicNormTypes()? Included normalizations are another feature that can differ between .hic files based on processing. The default norm value of readHic is "KR", but it is possible this may not be compatible with your file. If you don't have any normalizations, you can try "NONE".
3. Are you reading the data in at an appropriate resolution, and is that resolution available to you (by checking with strawr::readHicBpResolutions())? I notice that the the region you are reading in is 5000 bp, which is actually quite small and will require a small resolution. Are resolutions smaller than 5000 bp processed in your file?
4. Is your file downloaded and available to your R session? I have had issues with things like Dropbox in the past when the file is only available in the cloud and the function cannot read it.

If none of these troubleshooting tips get you any data, it is also a valid possibility that there is no data in this region. As I mentioned above, this is a small region, and even if you read in the data with something like a 5000 bp resolution, you would only get one pixel of information. If this pixel happens to have no data, then you truly have no data for that region.

I hope these tips help!

Best, Nicole

Hi, Nicole, thanks so much for your detailed and patient response. By checking strawr::readHicChroms(), the chromname is 1,2,3.... How could I change the chrom name in the .hic file to chr1, chr2... Thanks!

[nekramer]

Unfortunately, at this point in the processing stage you cannot easily change the names of the chromosomes in your .hic file itself. This usually goes back to whatever reference and software you used to process your data.

Based on the output of strawr::readHicChroms(), your original input of chr="1" should work. Can you try getting data for the whole chromosome to see if that is working? For this you would just use readHic(file, chrom = "1", chromstart = NULL, chromend = NULL).

Best,
Nicole