Simpler mode of inheritance of transcriptional variation in male Drosophila melanogaster

Wayne et al. 10.1073/pnas.0705441104.

Supporting Information

Files in this Data Supplement:

SI Dataset 1
SI Table 2
SI Dataset 2
SI Dataset 3
SI Table 3
SI Figure 4
SI Table 4

Fig. 4. Diagram of block strategy for partially balanced incomplete block design for pooling of RNA samples. Crosses were distributed across subblocks because of the size of the experiment; four subblocks for each of two blocks were performed with half the total number of crosses reared in each subblock.

SI Dataset 1

Dataset 1. Full results of analyses (n = 9345). Each gene is indexed by a unique probeuid. In this table gene, full name, and chromosome are given for information. More complete annotation is located in SI Data Set 2. F tests are labeled with the convention f_<test name>_<sex>, nominal P values from the F test are labeled probf_<test>_<sex>. Significance is indicated as probf_<test>_<sex>. "sex_signficant" is an indicator of whether the amount of transcript is significantly different between the two sexes, and if so, "bias_toward" indicates in which direction (e.g., greater in males indicated by male); if not, cells are empty. "xsome_id" is an indicator, whereas "autosome" represents chromosome 2 or 3; 4 represents the 4th or dot chromosome; and X and Y represent the two sex chromosomes. "on_group" is an indicator of the detection of the transcript with values "both" for detected in both sexes, "male" if detected only in males and "fem". Significance is presented for each component of variance for each sex separately, for an FDR of 0.10. For example, "gca_male" would have a "y" if GCA in males were significant at an FDR of 0.10, or an "n" if not signficant. "any_sig_male" indicates whether or not any components of variance were significant in males; similarly, "any_sig_female" indicates whether or not any components of variance were significant in females. For these indicator variables, if genes are sex-limited (on in only one sex), cells are empty for the sex in which expression is not detected. "simple" applies only to genes on in both sexes, and indicates genes which were strictly additive in males and strictly nonadditive in females with y; other genes on in both sexes are indicated with n. The four tests (GCA,SCA,RGCA,RSCA) in that order are combined to give a pattern of significance as pattern_female and pattern_male. The value "y" indicates significance while the value "n" indicates nonsignificance at FDR 0.10 for that particular test. Thus "ynnn" is significant for GCA alone.

SI Dataset 2

Dataset 2. Results of analyses for autosomal (A; chromosomes 2 and 3 only) and X-linked genes only where transcript is detected in both sexes (n = 8,607). Each gene is indexed by a unique probeuid. In this table gene, full name, chromosome, and cytological position are given for information. Significance at an FDR threshold of 0.10 is indicated as probf_<test>_<sex>. F tests are labeled with the convention f_<test name>_<sex>, nominal P values from the F test are labeled probf_<test>_<sex>. Significance is indicated as probf_<test>_<sex>. If expression is significantly different between the sexes , then "sex_significant" is has a "y" indicator and "bias_toward" is an indicator of whether the amount of transcript is greater in males or females. "xsome_id" is an indicator where "autosome" represents chromosome 2 or 3; 4 represents the 4^th or dot chromosome; and X represents the X chromosome. Significance is presented for each component of variance for each sex separately, for an FDR of 0.10. For example, "gca_male" would have a "y" if GCA in males were significant at an FDR of 0.10, or an "n" if not significant. "any_sig_male" indicates whether or not any components of variance were significant in males; similarly, "any_sig_female" indicates whether or not any components of variance were significant in females. The four tests (GCA,SCA,RGCA,RSCA) in that order are combined to give a pattern of significance as pattern_female and pattern_male. The value "y" indicates significance while the value "n" indicates non_significance at FDR 0.10 for that particular test. Thus "ynnn" is significant for GCA alone.

SI Dataset 3

Dataset 3. Abbreviated results plus full annotation information (n = 9,345). Each gene is indexed by a unique probeuid and the probe sequence is given. In this table gene, full name, and chromosome, and cytological position, Molecular_Functions, PhysicalMapEnd, PhysicalMapStart, Properties, Protein_Domains, Recombination_Map, Synonyms, GameFileID, Full_Name, Gadfly. Annotation is from Flybase version 4.7 and was extracted using the program developed previously (McIntyre et al. 2006). In addition, relevant lists of significant results from Parisi et al. 2004 are identified by merging the gene names from the Affymetrix chip with the gene names identified here; presence on any of the lists is an indicator variable with the column header name corresponding to the nomenclature in the original paper (ovaries_38, testes_39, soma_40, male_biased_41, female_biased_42). Similarly, significant results from Mackay et al. 2005 are indicated by the name of the supplementary table in that paper (M, MF, MFC, F, C, CF; note that results with MF are the ones presented in our paper). If expression is significantly different between the sexes, then "sex_significant" is has a "y" indicator and "bias_toward" is an indicator of whether the amount of transcript is greater in males or females. The column "autosome" indicates if a gene is on chromosome 2 or 3, in which case the indicator is 1; if it is not (i.e. it is located on the X, Y, or the 4^th, the indicator is 0. "xsome_id" is an indicator where "autosome" represents chromosome 2 or 3; 4 represents the 4th or dot chromosome; and X represents the sex chromosome. "on_group" is an indicator of the detection of the transcript with values "both" for detected in both sexes, "male" if detected only in males and "fem" if detected only in females. Significance is presented for each component of variance for each sex separately, for an FDR of 0.10. For example, "gca_male" would have a "y" if GCA in males were significant at an FDR of 0.10, or an "n" if not. "any_sig_male" indicates whether or not any components of variance were significant in males; similarly, "any_sig_female" indicates whether or not any components of variance were significant in females. For these indicator variables, if genes are sex-limited (on in only one sex), cells are empty for the sex in which the gene is not expressed. "simple" applies only to genes on in both sexes, and indicates genes which were strictly additive in males and strictly nonadditive in females with y; other genes on in both sexes are indicated with n. The four tests (GCA,SCA,RGCA,RSCA) in that order are combined to give a pattern of significance as pattern_female and pattern_male. The value "y" indicates significance while the value "n" indicates nonsignificance at FDR 0.10 for that particular test. Thus "ynnn" is significant for GCA alone.

SI Table 2

Table 2. For genes expressed in both sexes (n = 8,607). Counts of genes for different combinations of modes of inheritance, for both sexes for all genes (left section of table). The total given for each effect (i.e., GCA only) is for the total number of genes showing this effect in either sex. On the right side of the table are the counts for effects, for females only, for genes with any evidence for additive variation in males (right section of table; 1,570 genes with significant GCA, RGCA, or any combination of these modes of inheritance in males).

SI Table 3

Table 3. Estimates of variance components. Each gene is indexed by a unique probeuid. In this table, gene, full name, and chromosome are given for information. More complete annotation is located in SI Data Set 2. If expression is significantly different between the sexes, then "sex_significant" is has a "y" indicator and "bias_toward" is an indicator of whether the amount of transcript is greater in males or females. "xsome_id" is an indicator where "autosome" represents chromosome 2 or 3; 4 represents the 4th or dot chromosome; and X represents the sex chromosome. Estimates are REML and named as follows: <sex>_est_<effect>. Heritability is estimated (Lynch and Walsh 1998) for each sex separately, and listed as <sex>_heritability. CV_A is estimated (Houle 1992) and listed as CV_A,_<sex>, Genetic correlation for those genes with GCA present in both sexes along with the standard error was calculated and is listed as RG and SERG respectively (Holland 2006).

SI Table 4

Table 4. All possible modes of inheritance are listed, both within loci (additivity and dominance) and between loci (additive x additive, additive x dominance, and dominance x dominance epistasis) to indicate terms impossible for hemizygous males, but possible for diploid females. Letters X and A indicate effects coming from the X or the autosomes, respectively. Possible terms given the simple models of expression defined below are presented for expression from X linked genes for males and females, and then for expression from autosomal genes, for males and females. Sex differences in possible modes of inheritance for gene expression are summarized in the bolded rows. The terms listed as sex differences are possible in females but not in males due to hemizygosity of the X in males.