How to find corresponding nucleotide locations (i.e. 123682765-123683049) from a list of chromosomal regions (i.e. 5p14-15, 5q13-15, 5q31-32, etc)

Question

[cuib]

I'm new to sequence analysis, so I might have posted a redundant question. If so, please refer me to the correct place. From my initial search, I couldn't find answers to my question.

Is there a computational way (I prefer to use R, but if there's any more useful tool, please feel free to suggest!) to find corresponding nucleotide locations (i.e. 123682765-123683049) from a list of chromosomal regions (i.e. 5p14-15, 5q13-15, 5q31-32, etc)?

My input data would be: 5p14-15, 5q13-15, 5q31-32, etc

I'd like to get result in a dataframe format: 1st column listing nucleotide start location and 2nd column would be listing nucleotide end location.

Also, I'm doing ATACseq analysis, and if you know any beginner friendly learning materials/videos, I'd love to learn more.

Thank you.

Answer 1

[Ram]

The term you're looking for is "cytoBand". That's what the 5p15 etc. notations are called.

Check out this post: How To Obtain Chromosome Locus From Coordinates

For each input such as 5p15-16, split by - then prefix the \d+[pq] part of the first sub-string to the second sub-string. You can then pick the start position corresponding to the first sub-string and the end position corresponding to the second sub-string, and those would be your start and end positions.

[cuib]

Hi, thank you so much for your link and the name "CytoBand"! It's really useful. I was able to find useful questions such as following.

Cytogenic Location To Genome Coordinates In R

Genomic coordinates for Cytogenetic bands with R

I started using RStudio's Terminal. But I'm unfamiliar with it. Is there a stepwise instruction somewhere?

Should I use the code below? Is this what you mean by \d+[pq]?

curl -s "http://hgdownload.cse.ucsc.edu/goldenPath/hg18/database/cytoBand.txt.gz" | gunzip  -c

Or

mysql --user=genome --host=genome-mysql.cse.ucsc.edu -A -D hg19  -e "select chrom, min(chromStart), max(chromEnd) from cytoBand where name like 'q36%' group by chrom;"

Should hg19 be hg38 (what we used)? Also, above command gives "mysql: command not found"

Thank you again for your help

[Ram]

These questions are a lot more basic than the methods question you asked at the beginning. It looks like you're going to need to install mysql and learn a bit of R (and some regular expressions), and I cannot help you with that. If you're not familiar with mysql and R, please involve someone near you who can help you with that.

[cuib]

Thank you for your reply. I'm familiar with R and regular expression in statistical context and have learned SQL and HiveQL basics, but not mysql and terminal in a genomics context. I want to learn more about this process for sure. What kind of class would you suggest for this kind of process? I personally have a limited access to people who are knowledgeable of what I want to learn. Thank you.

[Ram]

If you know R, regex and SQL in any context, they can be applied here. It's the data that is different so you should be fine.