Supplementary Table S1. Seawater dissolved inorganic carbon characteristics in the LC and HC treatments, before and after partial renewal of the medium in semi-continuous cultures. Total inorganic carbon (DIC), pH, salinity, nutrient concentrations, and temperature were used to derive all other parameters using a CO2 system analyzing software (CO2SYS).
DIC, total dissolved inorganic carbon; TA, total alkalinity. Data are shown as the means ± SD (n=3).
Supplementary Table S2. Sequencing data quality and sequence read alignment analysis.
Raw_reads, original reads; Raw_bases, number of bases in the original sequencing data; Clean_reads, the number of clean reads obtained after filtering; Clean_bases, the number of bases in clean reads after filtering; valid_base, valid base percentage; Q30, the percentage of bases with a Qphred value 〉 30 as a percentage of the total number of bases; GC, total G + C content in the clean bases as a percentage of the total number of bases. For read mapping analyses: Total reads, the number of reads after sequence filtering (e.g., the clean reads); Total mapped, the number of sequences mapped to the genome; Multiple mapped, the number of sequences with multiple alignment positions on the reference genome; Uniquely mapped, the number of sequences with unique alignment positions on the reference genome; Splice reads: uniquely mapped reads that were segmentally aligned two exons (also known as Junction reads) where non-splice reads show the sequences aligned to the exon with reads mapped in proper pairs and the number of sequences on the double-ended alignment.
Supplementary Table S3. RT-qPCR primers and amplification results.
Supplementary Table S4. Gene identifiers, abbreviations, and annotations for 115 EPS metabolism-related genes.
Supplementary Figure 1. Hierarchical clustering of expression values (mean fragments per kilobase of transcript per million mapped reads; FPKM) among samples for 194 genes encoding carbohydrate-active enzymes in the Phaeodactylum tricornutum genome (http://protists.ensembl.org/Phaeodactylum_tricornutum/Info/Index). The enzymes were identified based on the CArbohydrate-Active enZYmes (CAZy) database. The color scale ranges from saturated red indicating highly expressed genes to saturated blue indicating weakly expressed genes, while white indicates relatively moderate expression. The heatmap was generated using the ComplexHeatmap program (Gu et al., 2016) using k-means clustering with a one minus Pearson correlation distance metric and average linkage method to cluster genes into a dendrogram.
File format; File name; File size; Uniform resource locator/link to file
text/tab-separated-values, 20 data points