GEO Accession viewer

NCBI > GEO > Accession Display

Not logged in | Login

GEO help: Mouse over screen elements for information.

Sample GSM1940476

Query DataSets for GSM1940476

Status

Public on Dec 20, 2015

Title

LacI tile 5 no selection

Sample type

SRA

Source name

plasmid

Organism

Escherichia coli

Characteristics

strain: K12 MG1655 EcNR2
sequenced molecule: plasmid DNA

Treatment protocol

The lacI gene variant libraries were constructed by cloning oligonucleotides encoding the desired mutations into plasmid pSC101_lacI_specR amplified by PCR with primers (Supplementary Table 6; Taylor, et al. Nature Methods. 2015) to appropriately linearize, add BsaI recognition sequence (5’-GGTCTCN), and remove the wild type lacI coding sequence segment to be replaced.Constructing the single residue replacement library involved replacing lacI codons from 3–359 with one missense codon encoding each of the remaining 19 natural amino acids (6,802 sequences). Oligos encoding these mutations were synthesized on a B3 Synthesizer (CustomArray, Inc., Bothell, WA), and were organized into the following tiles spanning the lacI gene: (#1)3–38, (#2)39–74, (#3)75–110, (#4)111–146, (#5)147–182, (#6)183–218, (#7)219–254, (#8)255–290, (#9)291–326, (#10)327–359.

Growth protocol

For library transformations, the screening strain was made electrocompetent by harvesting early log-phase cells (10 mL per transformation at OD600 nm = 0.15–0.25), removing salt through two washes with ice cold 10% glycerol, and resuspended in 50 μL cold 10% glycerol. Ten ng of library plasmid were electroporated into the competent cells, which were recovered for 1 hr in 1 mL SOC medium. To estimate the number of transformants, 1 μL recovered cells were plated on selective LB spectinomycin medium, and the remainder of the recovered cells were added to 10 mL of LB spectinomycin medium and selected overnight. Screening strain cells expressing LacI variants that do not bind to operator DNA constitutively express tolC and gfp; these were eliminated through negative selection by overnight selection with colicin E1 protein. We added five μL of saturated library transformation culture to 150 μL LB spectinomycin medium supplemented with tenfold serial dilutions of purified colicin E1 protein (2.73 mg/mL), in the range from 1:100 to 1:1,000,000; a control population of the same library was grown overnight without colicin E1. Enrichment of DNA-bound lacI variants was verified by flow cytometry the next day by measuring the fraction of GFP+ cells in the colicin E1 incubated and control populations. A 1:100,000 dilution of colicin E1 (20.7 ng/mL) was generally found to be optimal.

Extracted molecule

other

Extraction protocol

Plasmids were prepped from each library using a Spin Miniprep Kit (Qiagen #27104). A nested amplicon PCR strategy was used to append barcodes and sequencing adapters. In a first round of PCR, amplicon primers were used to amplify 300 bp regions of the lacI gene and add standard Illumina MiSeq sequencing primer binding sites. The first six bases to be read in sequencing read 1 were degenerate to increase complexity and aid in base calling. In a second round of PCR, barcoding primers were used to append barcodes and flow cell adapter sequences.
Amplicons were purified, quantified using Kapa Biosystems Library Quanitification kit (part number KK4824), and normalized. Libraries were sequenced on the Illumina MiSeq using MiSeq Reagent Kit v3, 600 cycle (part number MS-102-3003) according to the manufacturer’s protocol.

Library strategy

OTHER

Library source

genomic

Library selection

other

Instrument model

Illumina MiSeq

Data processing

Library strategy: DNA-seq
Basecalls performed using CASAVA version 1.8.2
Paired sequencing reads were collapsed and filtered for sequencing errors using FLASH v. 1.2.11 with a maximum overlap of 300
Collapsed reads were then translated in three frames, aligned to wild-type protein LacI with BLAT v.35 using default settings, and the best translated alignment by percentage match was retained. Protein sequences were then trimmed according to the amplicon/tile number and cloning flanking sequence.
Sequences with mis-matches in the fixed flanking sequences or different than the expected length (containing insertions or deletions) were discarded. Only sequences harboring a single amino-acid change found in either the pre- or post-selection were retained for further analysis. Protein sequences were then counted for pre- and post-selection, respectively, for each amplicon. A psuedo-count of one was added. Counts between pre- and post-selection were then quantile normalized using the normalize.quantiles function from preprocessCore in R v.3.1.1. Log2 fold-changes were computed from the ratio of pre-selection divided by post-selection quantile normalized counts.
genome build: available on series record in file lacI_reference_DNA.txt
Supplementary_files_format_and_content: Depletion values for amino-acid and position pairs; tabular format; log2(pre/post negative selection)

Submission date

Nov 13, 2015

Last update date

May 15, 2019

Contact name

Alexander Garruss

E-mail(s)

garruss@fas.harvard.edu

Organization name

Harvard Medical School

Department

Genetics