DENDRO.recalculate: Recalculate mutation profile for each cluster

In zhouzilu/DENDRO: Dna based EvolutioNary tree preDiction by scRna-seq technOlogy

Description

This function calculates the mutation profile for each cluster, after pooling cells within each cluster together. Such that, the cells within each cluster have same mutation profiles and the result is more robust. The estimation is based on a maximum likelihood approach. Loci with mutation observed in all or no subclones are removed.

Usage

DENDRO.recalculate(X, N, Info, DENDRO_label, cluster.name = NULL, top = NULL, epi = 0.001, m = 2)

Arguments

X	An matrix contains variants allele read counts across cell (column) and loci (row).
N	An matrix contains total allele read counts across cell (column) and loci (row).
Info	Mutation loci information. Matrix with row number same to number of loci and column number > 1. This is optional. Default NULL
DENDRO_label	An integer 1D vector that decide the labeled cluster in each cell. Estimated by DENDRO.dist and DENDRO.cluster.
cluster.name	You can label the name of the cluster after observing the cell composition. This is optional. Default NULL
top	Since many mutations may be observed, 'top' ask how many top mutation sites you want to selected for downstream analysis based on its marginal likelihood. This is optional. Default NULL
epi	Sequencing error and rare RNA editing combined rate. Default 0.001 according to Illunima.
m	The polidy in maximum likelihood mutation estimation. Default 2.

Value

X	An matrix contains variants allele read counts across subclones (column) and loci (row).
N	An matrix contains total allele read counts across subclones (column) and loci (row).
Z	An mutation indicator matrix (1 for mutation, 0 for normal) across subclones (column) and loci (row).
Info	Input Info after filtering
Z_cluster_lg	An likelihood matrix of mutation inference across subclones (column) and loci (row).

Author(s)

Zilu Zhou

References

Li, B., et al., A likelihood-based framework for variant calling and de novo mutation detection in families. PLoS Genet, 2012. 8(10): p. e1002944.

Examples

demo_qc = FilterCellMutation(demo$X,demo$N,demo$Z,demo$Info,demo$label)
dist = DENDRO.dist(demo_qc$X,demo_qc$N,demo_qc$Z,show.progress=FALSE)
cluster=DENDRO.cluster(dist,label=demo_qc$label)
icd = DENDRO.icd(dist,cluster)
DENDRO_label = cutree(cluster,3)
demo_cluster = DENDRO.recalculate(demo_qc$X,demo_qc$N, demo_qc$Info, demo_qc$DENDRO_label, cluster.name=c('Cluster3','Cluster2','Cluster1'))

zhouzilu/DENDRO documentation built on May 21, 2020, 8 p.m.