Computational strategies for the preconditioned conjugate gradient method applied to ssSNPBLUP, with an application to a multivariate maternal model

Table 1 Characteristics of different ssSNPBLUP systems

Characteristic	ssSNPBLUP_Liu (Plink)^a	ssSNPBLUP_Liu (DP)^b	ssSNPBLUP_MS^c
Number of iterations	3,358	3,359	6,334
Smallest eigenvalue	\(2.304*10^{-6}\)	\(2.304*10^{-6}\)	\(1.989*10^{-6}\)
Largest eigenvalue	3.813	3.813	5.194
Spectral condition number	\(1.655*10^{6}\)	\(1.655*10^{6}\)	\(2.612*10^{6}\)
Software peak memory (MB)^d	18,120.7	89,615.7	27,780.3

^assSNPBLUP model proposed by Liu et al. [9] and using the Plink 1 binary form; or ^b using double precision reals; ^cssSNPBLUP model proposed by Mantysaari and Stranden [11] and using the Plink 1 binary form; ^dThe software peak memory is defined as the peak resident size (VmHWM) obtained from the Linux /proc virtual file system

ISSN: 1297-9686