Features
Fragment Length (frag-length-bins, frag-length-intervals)
FinaleToolkit can calculate the fragment length distribution and summary statistics of cell-free DNA fragments genome-wide and across intervals.
Fragment Coverage (coverage)
FinaleToolkit can calculate fragment coverage, which refers to the number of reads mapped to a specific region of the genome. This coverage can be reported in two ways:
- Normalized Coverage: Adjusted by the total number of reads in the sample.
- Raw Coverage: The raw count of mapped reads.
Windowed Protection Score (WPS) (wps, adjust-wps)
FinaleToolkit can calculate the Windowed Protection Score (WPS), a metric introduced by Snyder et al., 2016.
WPS quantifies the protection or coverage of DNA fragments across specific genomic regions. It is calculated by:
- Counting the number of DNA fragments that fully span a defined window (typically 120 base pairs) centered around a genomic coordinate.
- Subtracting the number of fragments with endpoints falling within the same window.
This score is correlated with genomic features like nucleosome positioning, transcriptional start sites (TSS), and DNase I hypersensitive sites (DHSs).
FinaleToolkit can calculate the DELFI score, a metric introduced by Cristiano et al., 2019.
DELFI identifies abnormalities in cfDNA fragmentation patterns and was originally used to:
- Categorize patients with cancer.
- Determine the tumor tissue of origin.
The DELFI score is calculated as the ratio between:
- GC%-corrected short fragment count.
- GC%-corrected long fragment count.
End Motifs (end-motifs, interval-end-motifs)
FinaleToolkit can calculate the frequency of end-motif k-mers in cfDNA fragments.
End motifs, specific sequences found at fragment ends, are associated with cleavage patterns and can provide insights into conditions or diseases.
Motif Diversity Score (MDS) (mds, interval-mds)
FinaleToolkit can calculate the Motif Diversity Score (MDS), a metric introduced by Jiang et al., 2020.
MDS quantifies the diversity of cfDNA end motifs using the normalized Shannon entropy of the categorical distribution of all possible end-motif k-mers.
Cleavage Profile (cleavage-profile)
FinaleToolkit can calculate the Cleavage Proportion, a metric introduced by Zhou et al., 2022.
The cleavage proportion is defined as:
- The number of fragment ends falling within a nucleotide location.
- Divided by the total number of fragment ends overlapping that nucleotide location.
This metric has been shown to correlate with DNA methylation, particularly at CpG sites.
FinaleToolkit