Pacific Biosciences of California Inc (PACB)

[Paulieme]

•
Successfully assembled “diploid”-like long-read data generating assembly with N50 > 2 Mbp using only PacBio data.
•
With enough PacBio data, one can start assembly from reads >10 kb: It reveals the diploid structure as quasi-linear chains in the string graph.
•
Toward an improved diploid assembler:
–
More rigorous theoretical framework for the diploid / polyploid graph traversing problem
–
Generate diploid consensus: need an efficient aligner to create string graphs from long reads
–
Phasing: combining SV discovery to SNP calling to “unzip” the bubbles
–
More testing cases:
-
Real biological diploids
-
Other diploid genome might have different structure
Arabidopsis 120 Mbp genome Two strains, Ler-0 & Col-0 sequenced separately
Transposons
45S rDNAs
Retrotransposons
Common repeat element lengths
Pre-assembled read length distribution
Acknowledgements
We thank Joe Ecker and Chongyun Lou (HHMI & Salk Institute) for providing the Col-0 DNA sample. We also thank Adam Phillippy (NBACC) and Michael Schatz (CSHL) for insightful discussions about assembly algorithms. --https://s3.amazonaws.com/files.pacb.com/pdf/String+Graph+Assembly+For+Diploid+Genomes+with+Long+Reads.pdf?utm_content=buffer8d567&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer