q-kim/laggingstrand

Exploring the relationships between DNA replication and mutation.

LaggingStrand

Exploring the relationships between DNA replication and mutation. The FIG* subdirectories contain the code used to generate figures in the manuscript Reijns et al, Nature 2015 (doi:10.1038/nature14183). Associated dependencies and code for data preparation are located in the main directory.

Queries

The code here was written to address specific questions, it has not been polished or optimised for reuse. It is provided as is, as a record of analyses undertaken. However, some aspects of the work do have potential for reuse and we are keen to encourage and support that reuse. Please email [email protected] if you have specific queries.

Data

Data used in this analysis and it's sources are summarised in the file dataSources.txt. EmRiboSeq data generated in this work is available from GEO under accession number GSE64521. The script ionProcessor.sh documents primary processing of the emRiboSeq fastq files and depends on the script riboReadEndBedToGidChrM.pl.

Yeast polymorphism, between-species relative substitution rate, genomic masking and transcript annotation are available as a pre-processed table (doi:dx.doi.org/10.7488/ds/204).

GID

Yeast genome analysis was performed in R using large data-frames with one row per nucleotide of the yeast genome. GID (gid) used throughout scripts denotes an integer index of the sacCer3 reference genome, effectively concatenating each of the chromosome reference sequences. GID is a 1-based coordinate system with no padding between chromosomes.

Memory requirements

The large data-frame approach to analysis used here requires substantial RAM for analysis. Loading all relevant data sources including the emRiboSeq data requires approximately 250GB RAM.