Info about input data
./reference_plm -f fixed.effects.nam.sorted.filtered.bin --num_fixed 26 -g ../iplant_data/projectedSNP.0010.dat --num_geno 103234 -p -r residuals.chr10.sorted.bin --num_r 4892 -c -v2 -e .1 -l 50
The fixed effects and residuals files are converted from their text equivalents in the statinf repo; the forward regression source code includes a utility, convertToBinary, that will perform this conversion.
For the SNP projection, I assume a git checkout of the iplant_data repo, running ./prepare in iplant_data, and projecting chromosome 10 via projectAll.m in Matlab (it should be simple to alter the script to project only #10).
The statinf repo is here:
http://svn.iplantcollaborative.org/g2p/statinf/GLM-Matlab/TACC_Dec_2010/data_maize_nam/
The iplant_data repo is here:
https://github.com/pyrovski/NAM-Maize-data-preparation-example
If you have access to xz, decompress this file (70 MB):
http://www.cs.arizona.edu/~pbailey/projectedSNP.0010.dat.xz
If not, try this one (272 MB):
http://www.cs.arizona.edu/~pbailey/projectedSNP.0010.dat.bz2