The advent of high throughput RNA sequencing technologies makes it possible to more directly assess the relative quantities of steady-state transcripts derived from maternally or paternallyinherited alleles. Similar approaches have successfully identified genes imprinted during different stages of mouse development. Here we assay imprinted gene expression by performing high throughput sequencing on poly-A selected RNA from embryo and endosperm derived from reciprocal crosses between two Arabidopsis thaliana accessions, Ler and Col-0. This strategy allowed us to distinguish transcripts derived from the maternally inherited or paternally inherited allele for a portion of expressed genes with Ler/Col-0 single nucleotide polymorphisms. We identified.200 genes with parentally biased expression patterns. Our experimental strategy is particularly robust for identifying paternally expressed imprinted genes, as transcripts derived from the paternal genome must come from one of the fertilization products. Over 40 genes are predominantly paternally expressed, including a large number of transcription factors and chromatin related proteins. Most of the imprinted genes we identify exhibit parentally biased expression rather than complete monoallelic expression, suggesting that dosage regulation is an important factor in gene imprinting. Although all mapped reads can be used to determine overall gene expression levels, only a fraction of the reads are informative for determining maternal or paternal allele expression �C those reads that overlap a Col/Ler SNP. Data from reciprocal crosses also allows parent-of-origin effects to be distinguished from R428 strain specific Niltubacin biases in gene expression. We used information on previously identified SNPs to identify reads as Col or Ler. We obtained between 1.56 and 1.96 million informative reads for each library and calculated the number of Col or Ler reads for each gene. Approximately 10,300 genes from each tissue had at least 15 informative reads when data from reciprocal crosses were combined. These genes exhibited a range of maternal to paternal expression ratios, but the average percent maternal transcripts for each gene in the embryo and endosperm was near the expectation of 50% and 67%, respectively, based on the genomic DNA content of each tissue. This is consistent with studies of maize endosperm, which show that expression is proportional to the genomic contribution of the parents for most genes. To identify imprinted genes we used the Storer-Kim method to test whether the proportion of maternal and paternal reads for each gene was significantly different from expectations , taking into account the allelespecific read counts from both reciprocal crosses in order to distinguish parent-of-origin effects from strain-specific effects. We initially considered genes with a p-value less than 0.01, identifying 148 potential imprinted genes in the embryo and 1437 in the endosperm. Five of the 11 previously identified imprinted genes passed this initial p-value cutoff, while the remaining known imprinted genes either lacked Ler/Col SNPs, had very few informative reads, or, in one case, fell just below the cutoff. Allele-specific expression data, RPKM values, and embryo-endosperm differential expression analysis for all genes is presented in Tables S2 and S3. There was very little evidence for expression or imprinting of transposable elements in either tissue. Upon closer examination of some of the most highly parentally biased genes, it was clear that transcripts from genes highly expressed in the seed coat were contaminating both the embryo and endosperm fractions. Some contamination from abundant seed coat transcripts is likely unavoidable given our method of seed dissection, which is performed in an aqueous solution on a glass slide. We used expression information from another seed gene expression set to further filter our data. Le et al. used laser capture microdissection to isolate tissue from embryo, endosperm, and seed coat in the Ws background and determined gene expression values using Affymetrix microarrays. The rank order correlation coefficient between our embryo or endosperm expression data and the LCMD gene expression data of tissues at the same stage of development was good.