 |
 |
GEO help: Mouse over screen elements for information. |
|
| Status |
Public on Sep 05, 2024 |
| Title |
WT_oligo_pool_dCP_screen_S2_ecdysone_rep1 |
| Sample type |
SRA |
| |
|
| Source name |
S2 cells + ecdysone
|
| Organism |
Drosophila melanogaster |
| Characteristics |
cell line: S2 cells + ecdysone
treatment: screen
|
| Treatment protocol |
For each biological replicate, 10^8 cells were resuspended in 400µL of a 1:1 dilution of HyClone MaxCyte electroporation buffer and serum-free SM, and electroporated with 20µg of the input libraries (see previous section) using the MaxCyte-STX system (‘Optimization 1’ protocol). For the induction of hormone-inducible enhancers, S2 cells were treated after 1h of recovery following electroporation, by adding 50μL of 10mg/mL with 20-Hydroxyecysone to 25mL flasks (~42μM final concentration).
|
| Growth protocol |
Drosophila cells were cultured at 27ºC and passaged every 2-3 days. S2 cells were cultured in Schneider’s Drosophila Medium (SM, Gibco, #21720-024) supplemented with 10% heat inactivated FBS (Sigma, #F7524) and 1% Penicillin/Streptomycin. OSC cells were cultured in M3 Insect Medium supplemented with 0.6 mg ml-1 glutathione, 10% FBS, 10 mU ml-1 insulin and 10% fly extract.
|
| Extracted molecule |
polyA RNA |
| Extraction protocol |
UMI-STARR-Seq was performed as previously described (DOI: https://doi.org/10.1002/cpmb.105)
UMI-STARR-Seq libraries were UMI-tagged and constructed as previously described (DOI: https://doi.org/10.1002/cpmb.105)
|
| |
|
| Library strategy |
OTHER |
| Library source |
transcriptomic |
| Library selection |
other |
| Instrument model |
NextSeq 2000 |
| |
|
| Description |
36bp
WT_oligo_pool_dCP_S2_ecdysone_FC_table.txt
|
| Data processing |
STARR-Seq paired-end reads were aligned to custom indexes containing the corresponding oligonucleotide sequences, using the align function from the Rsubread R package (version 2.12.2) with the following parameters: type= “dna”, unique= TRUE, maxMismatches= 3 (or = 5 for the mutant library). Only the pairs for which both reads could be aligned with expected orientations and positions were considered. Then, UMI sequences were retrieved, reads were collapsed as previously described (DOI: https://doi.org/10.1038/s41588-022-01048-5), and only the pairs with at least 5 reads in each of the two input replicates were considered for downstream analyses.
To calculate the activity of all pairs, we added one pseudocount read to UMI-collapsed counts and computed the log2 fold-change over input using DESeq2 (with at least two replicates). To normalize the samples between them and facilitate comparisons between different screens, we used the negative control sequence pair counts as scaling factors, so that the activities of negative control pairs are centered on zero. Before computing the individual activity of each candidate sequence, we first aimed at removing potential outlier negative control sequence, that might eventually show some activity in our screens. To do so, we assessed the activity of each negative control sequence by averaging its activity across all its observed combinations with other control sequences. Resulting activities were scale in R and only the 5’ and 3’ control sequences with a z-score value located between -1 and 1 were considered as valid, robust control sequences. Finally, these robust control sequences were used to compute the individual activity of each individual candidate sequence (which we refer to as “enhancers” for simplicity), by averaging their activities across all its observed combinations with at least 10 robust control sequences (otherwise, the individual candidate sequence was discarded). To classify an enhancer as active, we compared the activities of all its observed combinations with robust control sequences to the activities of control/control pairs using one-tailed Fisher’s exact tests (alternative= “greater”) followed by false discovery rate (FDR) multiple testing correction. Only the enhancer sequences with a log2 activity bigger than 1 and a FDR<0.05 were considered as active.
Assembly: dm3
Supplementary files format and content: For each replicate per condition, "umi_counts.txt" file contains the total ("total_counts" column) umi-collapsed counts ("umi_counts" column) for each combination of 5' ("L" column) and 3' ("R" column) enhancer IDs. Correspondance between IDs and enhancer sequences are available in Supplementary Tables 2,9,11.
Supplementary files format and content: For each oligo pool, the "FC_table.txt" file contains the log2 activity ("log2FoldChange" column) and the corresponding adjusted p.values ("padj" column) for each combination of 5' ("L" column) and 3' ("R" column) enhancer IDs (see Supplementary Tables 2, 9 and 11). The "ctlL" and "ctlR" specify whether 5' and 3' sequences were considered as robust controls, respectively. 5' and 3' individual activities are reported in the "indL" and "indR" columns, respectively, with "padjL" and "padjR" columns containing FDR-corrected p.values (only the enhancers with an individual acitivity>1 and a FDR<0.05 are considered active).
Library strategy: STARR-Seq
|
| |
|
| Submission date |
Jun 13, 2024 |
| Last update date |
Sep 05, 2024 |
| Contact name |
Alexander Stark |
| E-mail(s) |
stark@starklab.org
|
| Organization name |
The Research Institute of Molecular Pathology (IMP)
|
| Lab |
Stark Lab
|
| Street address |
Campus-Vienna-Biocenter 1
|
| City |
Vienna |
| ZIP/Postal code |
1030 |
| Country |
Austria |
| |
|
| Platform ID |
GPL30203 |
| Series (1) |
| GSE245033 |
Developmental and housekeeping transcriptional programs display distinct modes of enhancer-enhancer cooperativity in Drosophila |
|
| Relations |
| BioSample |
SAMN41813205 |
| SRA |
SRX24900736 |
| Supplementary file |
Size |
Download |
File type/resource |
| GSM8326095_WT_oligo_pool_dCP_screen_S2_ecdysone_rep1_umi_counts.txt.gz |
6.3 Mb |
(ftp)(http) |
TXT |
SRA Run Selector |
| Raw data are available in SRA |
|
|
|
|
 |
