PMC

Cell-cycle recovery from stress. (A) Dynamic regulatory map derived for heat shock. As part of the recovery process, several genes regulated by Swi4, Swi6, Mbp1, and Fkh2 increase their expression level, reaching a level higher than their original (nontreated) values. Many of these genes are G1 genes (P-value <4 × 10⁻²⁰ based on a set of 300 G1 genes from ). (B) Expression profiles of genes assigned to the bifurcation event at the 40 min time point. Blue profiles represent genes assigned to the upper path of this bifurcation node. This path was controlled by the above cell-cycle TFs. Note that the expression of many of these genes rises above their initial expression value starting at the 60 min time point. (C) Budding index before and after heat shock. Cells are initially arrested and as predicted by DREM the percentage of G1 cells peaks starting at the 60 min time point. Following that peak, cells resume their cell-cycle activity in a more synchronized manner.

Dynamic regulatory map and static network for yeast response to AA starvation. (A) Dynamic map of yeast response to AA starvation using static input from condition-specific binding experiments and time-series expression data. TFs with split score below 0.001 appear next to the split they regulate, in ranked order of scores. Nodes in the graph represent hidden states. The area of a node is proportional to the s.d. of the expression of the genes assigned to that node. Green nodes represent split nodes. Many of the TFs were correctly assigned to the time points they are known to regulate. For example, Gcn4, which is a known master regulator of AA starvation response, is correctly assigned to the first split. Many of the TFs assigned to the second split regulate specific AA biosynthesis pathways. (B) Dynamic map of yeast response to AA starvation using input from both condition- and non-condition-specific ChIP-chip experiments. Several additional TFs not profiled with a condition-specific ChIP-chip experiment under the AA condition were determined to be participating in the response and recovery processes. These included Abf1, Swi4, Mbp1, and Ino4. In addition to identifying these TFs as potential participants in the response, DREM also identifies their time of influence. (C) Static regulatory graph for AA starvation. Nodes correspond to genes or TFs. An edge implies that the TF binds the gene with a P-value <0.005 in an AA starvation ChIP-chip experiment. Blue edges represent interactions between TFs. Whereas some properties of the networks can be derived from the static representation, many of the dynamic aspects of the system are lost when not using the time-series data.

Model overview. (A) Plots of time series expression profiles generated to illustrate the model. (B) Static TF-DNA binding data—DREM integrates TF-gene regulatory relationships derived from ChIP-chip or motif data with the time series expression data. For this example a majority of the pink genes in (A) are regulated by TF A, the blue genes by TF B and the red genes by TF C and D. (C) The model structure inferred by DREM for the data in (A) and (B). After the model is derived genes are assigned to their most likely paths based on their expression profile as well as on the set of TFs that regulate them. TF labels appear on some of the paths out of splits. (D) IOHMM model—each state has a Gaussian emission distribution for the expression values and the transition probabilities for a gene depend on the set of TFs that regulates it. A logistic regression classifier (Krishnapuram et al, 2005) maps the set of regulating TFs to transition probabilities. The classifiers are denoted by question marks in the figure. Example transition probabilities are given for a gene which is regulated by TF B. These probabilities are greater for the states with distributions similar to those of TF B regulated genes. The TF information also affects the structure of the resulting IOHMM model. Based on this information some splits can be added and some splits are removed from the model.

The role of Ino4 in regulating response to AA starvation. (A) Expression profiles of 13 genes in AA starvation that were assigned to the brown path in . These 13 genes were all bound by Ino4 in a ChIP-chip experiment in YPD media with a P-value <0.005 and have an evolutionarily conserved Ino4 motif. It was predicted by DREM that Ino4 was activating these and other genes starting around 2 h (see also ). (B) Occupancy rates of Ino4 in the promoter region of four genes regulated by Ino4, before and at 4 h after AA starvation. For three of these four genes, the Ino4 promoter occupancy rates were at least two-fold higher following AA starvation than in synthetic complete+D-glucose (SCD) media before AA starvation. (C) Comparison of the number of genes bound by Ino4 before and 4 h after AA starvation using a whole-genome binding experiment. We compared the lists using two different P-value cutoffs (0.001 and 0.005). Genes were counted if they are bound at the appropriate P-value in at least one of the two repeats. At the 0.001 P-value cutoff, there is almost a six-fold enrichment for Ino4-bound genes 4 h after AA starvation. (D) Comparison of binding P-values for genes assigned to the main path determined by DREM to be regulated by Ino4 in one of the repeats (see also ). The plots are the negative log base 10 of the binding P-value for genes that were bound with a P-value <0.005 in one or more of the Ino4 binding experiments and are on the identified Ino4 response path. The horizontal and vertical lines represent a P-value significance of 0.005. Anything to the right of the vertical line is significant under normal growth conditions. Anything above the horizontal line is significant in the AA starvation experiment. Anything above the diagonal line is more significant in the AA starvation experiment. This plot indicates that these genes were bound more significantly in AA starvation conditions than SCD conditions.

Distribution of binding overlap between YPD media and stress condition for different sets of TFs. We group TF into two sets. The first subset contains the TFs assigned to first split points in AA starvation or heat that were profiled with a ChIP-chip experiment in the condition (8, primary TFs Cbf1, Gcn4, Fhl1, Rap1, and Sfp1 from AA starvation in and for heat Msn2 and Skn7 from and Hsf1 from ) and the secondary TFs assigned to second split point profiled with a ChIP-chip experiment in the condition (8, secondary TFs Arg81, Dal82, Gln3, Hap5, Met32, Met4, Rtg3, and Stp1 from AA starvation in ). (A) Percent overlap for each of these two sets when binding (under both stress and YPD media conditions) is determined using a 0.005 P-value cutoff. Note the difference between the distribution of overlap for primary and other TFs. Whereas the majority of TFs display a big difference in the set bound genes under stress and YPD media, many primary TFs bind to a large percentage of stress-regulated genes in YPD media as well. This difference is even bigger in (B) where we plot the overlap for the top 100 genes (ordered by their binding P-values) in each condition. Whereas most TFs (and most secondary TFs) drastically alter the subset of genes they regulate under stress, half of the primary factors bind to more than 50% of the same genes in both conditions. Although the binding strength may be different under stress, these results indicate that many of the primary pathways are maintained, in low levels, under YPD media as well. (C) Average expression levels for primary and secondary factors for the first two time points in the AA starvation and heat-shock experiments. Whereas the average expression levels for the secondary factors are much higher when compared to their untreated levels, the levels for the primary factors do not change significantly between the two conditions. (D) Comparison of binding P-values for Gcn4 in the MMS condition at 15 and 60 min. Points above the diagonal correspond to genes that were bound more significantly at 15 min than at 60 min. As can be seen from the plot, a substantial majority of genes were bound more significantly at 15 min than at 60 min. Points to the right of the vertical had a P-value <0.005 in MMS at 60 min, whereas points above the horizontal line had a P-value <0.005 in MMS at 15 min.