Hi,

I am trying to index the human genome with buildindex from the Rsubread package. However, I receive a fatal error with macOSX Mojave.

The genome has been downloaded from the UCSC server and all chromosomes are combined to a single fastA file. Repeats are softmasked.

# path to genome file
ref = normalizePath(path ="~/Desktop/work/fusion/work/alignments/storrage/hsa.fa")
# building the index
buildindex(basename = "hsa_index", reference = ref) 

//================================= Running ==================================\\
||                                                                            ||
|| Check the integrity of provided reference sequences ...                    ||
|| No format issues were found                                                ||
|| Scan uninformative subreads in reference sequences ...                     ||
|| 676060 uninformative subreads were found.                                  ||
|| These subreads were excluded from index building.                          ||
|| Build the index...                                                         ||
||    8%,  24 mins elapsed, rate=178.8k bps/s                                 ||
||   16%,  38 mins elapsed, rate=223.6k bps/s                                 ||
||   24%,  55 mins elapsed, rate=232.0k bps/s                                 ||
||   33%,  75 mins elapsed, rate=227.1k bps/s                                 ||
||   41%, 103 mins elapsed, rate=208.0k bps/s                                 ||
Killed: 9

About the Mac

MacBook Pro (Retina, 13-inch, Late 2013) Processor 2,8 GHz Intel Core i7 Memory 16 GB 1600 MHz DDR3

As of now I am not clear how to handle this problem, advice is very welcome.

Carsten

Your index building process was likely to be terminated by the system due to large amount of memory used by buildindex for building a full index for human genome.

You may use the following command to build a gapped index to save memory. This will result in an increase in read mapping time but it is still fast.

buildindex(basename = "hsa_index", reference = ref, gappedIndex=TRUE)

To build a full index for human/mouse genome, buildindex function will need 15GB of memory. Mapping functions in Rsubread including align and subjunc will need 17.8GB of memory to perform read alignment using this full index. This enables fastest mapping speed, but it is recommended for use on a unix server due to relatively large memory use.

To use Rsubread in a personal computer that typically has 16GB of memory or less, a gapped index is recommended. buildindex only needs 5.7GB of memory to build a gapped index for human/mouse genome. align will need 8.2GB of memory for mapping with this index and subjunc will need 8.8GB of memory.

Memory use for index building and read mapping can be further reduced by building a split index using the indexSplit and memory options in buildindex.