ggirelli / gpseq-radical

R script for fast and streamlined Radiality Calculation from GPSeq data.

GPSeq-RadiCal

R script for fast and streamlined Radiality Calculation from GPSeq data. GPSeq-RadiCal aims to provide the same functionalities as the gpseqc Python3 package, but with much lower memory usage and faster computation times.

Installation

• Tested on R v3.6.3.
• Required packages: argparser, data.table, logging, outliers, pbapply, rtracklayer.

The script checks automatically for the required packages.
When packages are missing, it provides the code to install the missing ones, one by one.

Usage

First, prepare a tabulation-separated metadata file with four columns:

• exid: sequencing run ID.
• cond: condition description (e.g., time and/or concentration).
• libid: library ID.
• fpath: full absolute path to bed file. Bed files can be gzipped (recommended).

The bed files should be reported in order of condition strength, with the top rows being weaker than bottom ones. Also, the first line should contain the column headers. An example metadata file can be found here.

Then, to run with default parameters, execute the following command:

./gpseq-radical.R example_meta.tsv output_folder

Replacing example_meta.tsv with your metadata file path, and output_folder with the path where the script should write the output. Note that the specified output folder must not already exist.

To access the script help page, run the following:

./gpseq-radical.R -h

Differences from gpseqc

• Allows for normalization in a chromosome-wise fashion too, instead of only library-wise.
• Normalization factors are calculated after outlier removal, not before.
• Masking is performed before binning, but after the normalization factors are calculated. In this way, reads from masked regions still count for normalization purposes, while they are masked out from chromosome-wide analysis.

Desired features

• Re-running on the same output folder to populate with additional resolutions.
• Additional centrality estimates (currently calculates only the one selected in the GPSeq study).

Contributing

We welcome any contributions to GPSeq-RadiCal. Please, refer to the contribution guidelines if this is your first time contributing! Also, check out our code of conduct.

License

MIT License
Copyright (c) 2020 Gabriele Girelli

References

• Girelli, Gabriele, et al. "GPSeq reveals the radial organization of chromatin in the cell nucleus." Nature Biotechnology (2020): 1-10. Genomic loci Positioning by Sequencing (GPSeq) original paper.

gpseq-radical's Issues

[REQUEST] Option to export bed outliers.

[BUG] Binning crashes with --site-domain universe

Should use seq_along instead of seq_len here:

GPSeq-RadiCal/gpseq-radical.R

Line 274 in 020c39a

tag=bbins2[1, tag], cid=seq_len(args$cond_cols), nsites=.N

[BUG] --ref-genome flag does not work

./GPSeq-RadiCal-0.0.9/gpseq-radical.R "$metadata" /home/${run_id}/test \
>        --chromosome-wide -b 1e6:1e5,5e5:5e4,1e5:1e4,5e4:5e4,25e3:25e3 \
>        --normalize-by lib \
>        --ref-genome mm10 --chrom-tag 19:X,Y \
>        --threads 6 --export-level 1 \
>        --site-domain universe --site-bed "$cutsite_path" \
>        --mask-bed "$mask_path"
2022-03-23 14:04:50 INFO::This log will be stored at '/home/WW319/test/gpseq-radical.log'.
2022-03-23 14:04:50 INFO::Created output folder '/home/WW319/test'.
2022-03-23 14:04:50 INFO::Exported input parameters to '/home/WW319/test/gpseq-radical.opts.rds'.
2022-03-23 14:04:50 INFO::Parsing metadata from '/home/B319_centrality_GPSeq_meta.tsv'.
2022-03-23 14:04:50 INFO::Storing metadata.
2022-03-23 14:04:50 INFO::Opening UCSC browser session...
2022-03-23 14:05:00 INFO::Querying UCSC for 'mm10' chromosome info...
Loading required package: BiocGenerics

Attaching package: ‘BiocGenerics’

The following objects are masked from ‘package:stats’:

    IQR, mad, sd, var, xtabs

The following objects are masked from ‘package:base’:

    anyDuplicated, append, as.data.frame, basename, cbind, colnames,
    dirname, do.call, duplicated, eval, evalq, Filter, Find, get, grep,
    grepl, intersect, is.unsorted, lapply, Map, mapply, match, mget,
    order, paste, pmax, pmax.int, pmin, pmin.int, Position, rank,
    rbind, Reduce, rownames, sapply, setdiff, sort, table, tapply,
    union, unique, unsplit, which.max, which.min

Loading required package: S4Vectors
Loading required package: stats4

Attaching package: ‘S4Vectors’

The following objects are masked from ‘package:base’:

    expand.grid, I, unname

Error in h(simpleError(msg, call)) :
  error in evaluating the argument 'object' in selecting a method for function 'getTable': Table 'chromInfo' is unavailable
Calls: <Anonymous> ... tableName<- -> tableName<- -> .local -> normArgTable
Execution halted