ESTIMATION OF CELL LINEAGE TREES BY MAXIMUM-LIKELIHOOD PHYLOGENETICS.

Area of Science:

• Developmental Biology
• Genomics
• Computational Biology

Background:

• CRISPR technology facilitates cell lineage tracing in multicellular organisms using genomic barcodes and insertion-deletion mutations.
• Current computational tools for reconstructing cell lineage trees are general-purpose and do not fully leverage the specific mutation process of CRISPR.
• Existing methods have limitations in inferring lineage relationships, particularly across parallel lineages.

Purpose of the Study:

• To develop a novel statistical model specifically for the CRISPR mutation process in genomic barcodes.
• To create an advanced computational procedure for accurate cell lineage tree reconstruction.
• To improve the inference of tree topology, branch lengths, and mutation parameters.

Main Methods:

• Proposed a statistical model for the CRISPR mutation process.
• Developed an iterative penalized maximum likelihood estimation procedure.
• Assumed barcode evolution follows a molecular clock model.

Main Results:

• The new method accurately estimates cell lineage tree topology, branch lengths, and mutation parameters.
• It infers relative ordering across parallel lineages, outperforming existing techniques.
• Application to transgenic zebrafish data recapitulated known developmental aspects and showed consistent results across samples.

Conclusions:

• The proposed statistical model and computational method significantly advance CRISPR-based cell lineage tracing.
• This approach provides a more robust and accurate reconstruction of cell lineage trees in complex organisms.
• The findings have implications for understanding developmental processes and genetic mosaicism.