Previously, we isolated Chlorella sp. HS2 (referred hereupon as HS2) from a local tidal rock pool and demonstrated its halotolerance and high biomass productivity under different salinity conditions. To further understand acclimation responses of this alga under high salinity stress, we performed transcriptome analysis of triplicated culture samples grown in freshwater and marine conditions at both exponential and stationary growth phases. The results indicated that the transcripts involved in photosynthesis, TCA, and Calvin cycles were downregulated, whereas the upregulation of DNA repair mechanisms and an ABCB subfamily of eukaryotic type ABC transporter was observed at high salinity condition. In addition, while key enzymes associated with glycolysis pathway and triacylglycerol (TAG) synthesis were determined to be upregulated from early growth phase, salinity stress seemed to reduce the carbohydrate content of harvested biomass from 45.6 dw% to 14.7 dw% and nearly triple the total lipid content from 26.0 dw% to 62.0 dw%. These results suggest that the reallocation of storage carbon toward lipids played a significant role in conferring the viability of this alga under high salinity stress by remediating high level of cellular stress partially resulted from ROS generated in oxygen‐evolving thylakoids as observed in a direct measure of photosystem activities.

Integrating -omics data with biological networks such as protein–protein interaction networks is a popular and useful approach to interpret expression changes of genes in changing conditions, and to identify relevant cellular pathways, active subnetworks or network communities. Yet, most -omics data integration tools are restricted to static networks and therefore cannot easily be used for analyzing time-series data. Determining regulations or exploring the network structure over time requires time-dependent networks which incorporate time as one component in their structure. Here, we present a method to project time-series data on sequential layers of a multilayer network, thus creating a temporal multilayer network (tMLN). We implemented this method as a Cytoscape app we named TimeNexus. TimeNexus allows to easily create, manage and visualize temporal multilayer networks starting from a combination of node and edge tables carrying the information on the temporal network structure. To allow further analysis of the tMLN, TimeNexus creates and passes on regular Cytoscape networks in form of static versions of the tMLN in three different ways: (i) over the entire set of layers, (ii) over two consecutive layers at a time, (iii) or on one single layer at a time. We combined TimeNexus with the Cytoscape apps PathLinker and AnatApp/ANAT to extract active subnetworks from tMLNs. To test the usability of our app, we applied TimeNexus together with PathLinker or ANAT on temporal expression data of the yeast cell cycle and were able to identify active subnetworks relevant for different cell cycle phases. We furthermore used TimeNexus on our own temporal expression data from a mouse pain assay inducing hindpaw inflammation and detected active subnetworks relevant for an inflammatory response to injury, including immune response, cell stress response and regulation of apoptosis. TimeNexus is freely available from the Cytoscape app store at https://apps.cytoscape.org/apps/TimeNexus.

1 stress suggest the overflow of acetyl-CoA from glycolysis and NADPH co-factor induces 2 high lipid accumulation and halotolerance in Chlorella sp. HS2 3 4 Abstract 32 Previously, we isolated Chlorella sp. HS2 (referred hereupon HS2) from a local tidal rock pool 33 and demonstrated its halotolerance and relatively high biomass productivity under different 34 salinity conditions. To further understand acclimation responses of this alga against high 35 salinity stress, we performed transcriptome analysis of triplicated culture samples grown in 36 freshwater and marine conditions at both exponential and stationary growth phases. De novo 37 assembly followed by differential expression analysis identified 5907 and 6783 differentially 38 expressed genes (DEGs) respectively at exponential and stationary phases from a total of 52770 39 transcripts, and the functional enrichment of DEGs with KEGG database resulted in 1445 40 KEGG Orthology (KO) groups with a defined differential expression. Specifically, the 41 transcripts involved in photosynthesis, TCA and Calvin cycles were downregulated, whereas 42 the upregulation of DNA repair mechanisms and an ABCB subfamily of eukaryotic type ABC 43 transporter was observed at high salinity condition. In addition, while key enzymes associated 44 with glycolysis pathway and triacylglycerol (TAG) synthesis were determined to be 45 upregulated from early growth phase, salinity stress seemed to reduce the carbohydrate content 46 of harvested biomass from 45.6 dw% to 14.7 dw% and nearly triple the total lipid content from 47 26.0 dw% to 62.0 dw%. These results suggest that the reallocation of storage carbon toward 48 lipids played a significant role in conferring the viability of this alga under high salinity stress, 49 most notably by remediating high level of cellular stress partially caused by ROS generated in 50 oxygen-evolving thylakoids. 51 52 Summary Statement 53 Redirection of storage carbon towards the synthesis of lipids played a critical role in conferring 54 the halotolerance of a Chlorella isolate by remediating excess oxidative stress experienced in 55 photosystems. 56 57

Integrating -omics data with biological networks such as protein-protein interaction networks is a popular and useful approach to interpret expression changes of genes in changing conditions, and to identify relevant cellular pathways, active subnetworks or network communities. Yet, most -omics data integration tools are restricted to static networks and therefore cannot easily be used for analyzing time-series data. Determining regulations or exploring the network structure over time requires time-dependent networks which incorporate time as one component in their structure. Here, we present a method to project time-series data on sequential layers of a multilayer network, thus creating a temporal multilayer network (tMLN). We implemented this method as a Cytoscape app we named TimeNexus. TimeNexus allows to easily create, manage and visualize temporal multilayer networks starting from a combination of node and edge tables carrying the information on the temporal network structure. To allow further analysis of the tMLN, TimeNexus creates and passes on regular Cytoscape networks in form of static versions of the tMLN in three different ways: i) over the entire set of layers, ii) over two consecutive layers at a time, iii) or on one single layer at a time. We combined TimeNexus with the Cytoscape apps PathLinker and AnatApp/ANAT to extract active subnetworks from tMLNs. To test the usability of our app, we applied TimeNexus together with PathLinker or ANAT on temporal expression data of the yeast cell cycle and were able to identify active subnetworks relevant for different cell cycle phases. We furthermore used TimeNexus on our own temporal expression data from a mouse pain assay inducing hindpaw inflammation and detected active subnetworks relevant for an inflammatory response to injury, including immune response, cell stress response and regulation of apoptosis. TimeNexus is freely available from the Cytoscape app store at https://apps.cytoscape.org/apps/TimeNexus.