My data is setup as in the MetaDE vignette:

•  gene_tstats & gene_pvals are matrices with rownames corresponding to gene names and moderated t-statistic / p-values for each study.

•  gene_tstats and gene_pvals have the same column names and dimensions (20917 * 3) and no NA values (although this shouldn't matter?)

data <- list(stat=gene_tstats, p=gene_pvals)
metade <- MetaDE.pvalue(data, meta.method='AW.OC')

I get this error:

Warning message:
In get.AW.OC(x$p, x$bp): there're no parametric results for Pearson's method,we will use simulation to estimate
       the p valuesWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrix

Error in xj[i]: invalid subscript type 'list'

If I use a subset of my data, there are no error messages:

data100 <- list(stat=gene_tstats[1:100,], p=gene_pvals[1:100,])
metade <- MetaDE.pvalue(data100, meta.method='AW.OC')

Warning message: In get.AW.OC(x$p, x$bp): there're no parametric results for Pearson's method,we will use simulation to estimate the p values

Larger subsets produce different warning messages but do return a result successfully:

data2000 <- list(stat=gene_tstats[1:2000,], p=gene_pvals[1:2000,])
metade <- MetaDE.pvalue(data2000, meta.method='AW.OC')

Warning message:
In get.AW.OC(x$p, x$bp): there're no parametric results for Pearson's method,we will use simulation to estimate
       the p valuesWarning message:
In matrix(Vb[, i], nr, np/nr): data length [5000] is not a sub-multiple or multiple of the number of rows [2000]Warning message:
In matrix(Vb[, i], nr, np/nr): data length [5000] is not a sub-multiple or multiple of the number of rows [2000]Warning message:
In matrix(Vb[, i], nr, np/nr): data length [5000] is not a sub-multiple or multiple of the number of rows [2000]Warning message:
In matrix(Vb[, i], nr, np/nr): data length [5000] is not a sub-multiple or multiple of the number of rows [2000]Warning message:
In matrix(Vb[, i], nr, np/nr): data length [5000] is not a sub-multiple or multiple of the number of rows [2000]Warning message:
In matrix(Vb[, i], nr, np/nr): data length [5000] is not a sub-multiple or multiple of the number of rows [2000]Warning message:
In matrix(Vb[, i], nr, np/nr): data length [5000] is not a sub-multiple or multiple of the number of rows [2000]

This only works up to a point:

data10000 <- list(stat=gene_tstats[1:10000,], p=gene_pvals[1:10000,])
metade <- MetaDE.pvalue(data10000, meta.method='AW.OC')

Warning message:
In get.AW.OC(x$p, x$bp): there're no parametric results for Pearson's method,we will use simulation to estimate
       the p valuesWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrixWarning message:
In matrix(Vb[, i], nr, np/nr): data length exceeds size of matrix

Error in xj[i]: invalid subscript type 'list'

I't seems to be something related to the number of genes I want to do a meta analysis for. Suggestions on how to resolve (other than to subset my data)?