Additional file 1:

Log2 expression values for the 445 candidate genes selected from our microarray analysis. This file contains log2 expression values (relative to time 0) for the 445 candidate genes selected from our microarray analysis. There are two sheets labeled 'Average_values' and 'Duplicate_experiments'. The 'Duplicate_experiments' sheet contains the values of two independent experiments (denoted _1 and _2 following the name of strain and time). The 'Average_values' sheet contains the data represented in Figure 2, corresponding to the average values of the two independent experiments mentioned above. Arrays are labeled with the relevant genotype of the strain and the time of sampling. Same color is used for all the arrays obtained with the same strain. The background context for all strains was bck2Δ P_MET3·CLN2. Except for strain cln3Δ, cells also had P_GAL1·CLN3 at the endogeneous CLN3 locus (wt stands for wild type).

Format: XLSX Size: 337KB Download file

Additional file 2:

Log likelihood scores for the 445 candidates analyzed in our study. Matrix containing the individual values assigned to each gene in all nine classifiers used in our model and the final score obtained (column SUM). Each sheet corresponds to one TF. 'PEAK TIME' evaluates the time of peak expression in the wild-type strain in our experiments. 'Value 20' wt' evaluates the value at 20 minutes in the wild-type strain whereas 'Av. value 40-60 wt' (only Swi4) corresponds to the average value at 40 and 60 minutes in the wild type. In 'Corr. wt/mbp1Δ' we assess the value for the correlation coefficient between the expression patterns in the wild type versus the mbp1Δ strain. 'max wt/max mbp1Δ' (only Mbp1) refers to the ratio between the maximum value in the wild-type series (20 to 80 minutes) and the maximum in the mbp1Δ mutant. Similarly, 'max wt_20-40/max swi4Δ' makes reference to the ratio between the maximum value at 20 or 40 minutes in the wild type and the maximum value in the swi4Δ background. For 'Mbp1 motifs' we evaluated whether the promoters of genes had at least one MCB consensus site (ACGCGT) within the first 200 bp upstream of the TSS or not. For SBF ('Swi4 motifs'), we examined the presence of at least one SCB consensus site (CRCGAA) located within 400 bp of the TSS. In 'Mbp1 binding' we evaluate TF binding data from four genome-wide studies that used ChIP-chip technology [15-18]. We considered the assignments proposed by Iyer et al. [15], and those TF-target interactions with a P-value ‹0.001 from the other three studies. Three conditions were assessed: none of the studies, only one study, and more than one study detected an interaction. The same applies to 'Swi4 binding' but interactions detected by Iyer et al. were considered more reliable and consequently given more weight (see Materials and methods for details). In 'cln3' and 'clb2', we analyzed the value of induction upon Cln3 or upon Clb2 overexpression in [3]. Finally, 'CC peak' assesses the time of peak expression during the mitotic cell cycle.

Format: XLSX Size: 129KB Download file

Additional file 3:

Full lists of predicted targets. This file contains the full lists of predicted targets for both factors, MBF and SBF (two sheets labeled correspondingly). Cell cycle peak was taken from [20,45]. References for publications where binding of TF or predictions of the gene as target have been reported are given in the 'TF binding' and 'Previous classifications' columns, respectively. For MBF, the number of ACGCG motifs in the first 200 bp upstream of the TSS and motifs beyond the first 200 bp (between parentheses) are given. For SBF, the number of CRCGAA motifs in the first 400 bp upstream of the TSS and motifs beyond the first 400 bp (between parentheses) are given. 'SGD description' refers to gene product description in the Saccharomyces Genome Database. 'Functional class' includes some modified functional classes according to the MIPS functional catalog (see Materials and methods for details).

Format: XLSX Size: 92KB Download file

Additional file 4:

Cell cycle distributions of predicted targets according to the timing of peak expression. This figure constitutes an expansion of Figure 4 of the paper. It shows the cell cycle distributions of predicted targets according to the timing of peak expression for a number of classifications: F, this study; B, [22]; H, [23]; BJ, [35]; T, [36]; W, [38]; G, [37]. Values on the x-axis are percentages of the whole duration of the cycle, as defined in [20]. Red, MBF targets; blue, SBF targets; green, both MBF and SBF targets; gray, our 445 candidates not classified eventually as MBF or SBF targets. Note that y-axis scales vary across classifications.

Format: PDF Size: 186KB Download file

This file can be viewed with: Adobe Acrobat Reader

Additional file 5:

Cell cycle distributions of predicted targets according to the timing of activation of expression from [39]. Letter and color keys as in Additional file 4.

Format: PDF Size: 165KB Download file

This file can be viewed with: Adobe Acrobat Reader

Additional file 6:

Cell cycle distributions of predicted targets according to the timing of deactivation of expression from [39]. Letter and color keys as in Additional file 4.

Format: PDF Size: 189KB Download file

This file can be viewed with: Adobe Acrobat Reader

Additional file 7:

The 648 genes considered as cell cycle regulated in this study. These genes appear among the top 800 ranked in at least three of five cell cycle studies [3,20,45,52,53]. The time of their peak expression is also shown [20,45].

Format: XLSX Size: 69KB Download file