NP-SMLR is a software for GpC methylation detection and nucleosome occupancy inference at single-molecule level. It takes the aligned signal-levels from Oxford Nanopore sequencing reads, and outputs the posterior probability of methylation on every GpC site together with the positions of all inferred nuclosomes on each molecule.

Dependencies

The current version has been tested in the Linux system with:

1. GCC (version 7.3.0);
2. Perl (version 5.16.3);
3. SAMtools (version 1.9);
4. BEDtools (version 2.28.0);
5. Boost (version 1.73.0).

Installation

You can download and compile the latest version (v1.0) as follows:

git clone https://github.com/imatrm/NP-SMLR.git
cd NP-SMLR
make

After compilation, three executable files (Likelihood, Detection and NclsPos) will be generated in the folder NP-SMLR/bin.

Please ensure that you have added

1. SAMtools
2. BEDtools

to your PATH.

Before running NP-SMLR

Before running NP-SMLR,

1. the reads have be aligned to reference genome; and
2. the alignments between event signals and reference genome should be obtained using nanopolish eventalign. Please make sure that the flag --print-names is set.

The tutorial of nanopolish eventalign can be found at https://nanopolish.readthedocs.io/en/latest/quickstart_eventalign.html.

Usage

The command is

./NP-SMLR.pl -b sorted_bam_file -e event_align_scale -o output_dir

• Parameters

-b   <STRING>   Name of the BAM file that records the alignment of reads. The BAM file must be sorted.
-g   <STRING>   Event alignment file generated by "nanopolish eventalign" (with flag --scale-events).
-t   <STRING>   Name of output folder.

• Example

The test example can be run using the command

./NP-SMLR.pl -b ./testdata/sort_test.bam -e ./testdata/eventalign_scale_test.txt -o output

The generated files are stored under output.

Output

The file ncls_pos.bed in the output folder records the coordinates of nucleosomes detected on each molecule. The columns represent:

Column 1: Chromosome ID;

Column 2: Start position of nucleosome;

Column 3: End position of nucleosome;

Column 4: Name of molecule (read);

Column 5: Quality score of alignment (obtained from BAM file);

Column 6: Strand on which the read is aligned.

The file detection.txt in the output folder records the methylation score of every GpC site on each molecule, which is essentially the posterior probability of methylation. The columns represent:

Column 1: Chromosome ID;

Column 2: Position of GpC site;

Column 3: Nucleotide sequence covering the GpC site (the 6-mers on this sequence are involved in the detection of GpC methylation);

Column 4: Name of molecule (read);

Column 5: Log likelihood of event-level with respect to negative control;

Column 6: Log likelihood of event-level with respect to positive control;

Column 7: Methylation score of GpC site (the posterior probability of methylation).

Current version

The version of the current release is v1.0.

Contact

Please contact wanganqi18@gmail.com for any question.

License

Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License

For details, please read NP-SMLR/License.

Citation

Yunhao Wang*, Anqi Wang*, Zujun Liu, Andrew L. Thurman, Linda S. Powers, Meng Zou, Yue Zhao, Adam Hefel, Yunyi Li, Joseph Zabner, Kin Fai Au. Single-molecule long-read sequencing reveals the chromatin basis of gene expression. Genome Research, 2019. doi: 10.1101/gr.251116.119 (*contributed equally)