 |
 |
GEO help: Mouse over screen elements for information. |
|
| Status |
Public on Dec 05, 2023 |
| Title |
HOXD13-Q325K |
| Sample type |
protein |
| |
|
| Source name |
IVT protein
|
| Organism |
Homo sapiens |
| Characteristics |
sample type: Variant
|
| Treatment protocol |
n/a/
|
| Growth protocol |
n/a
|
| Extracted molecule |
protein |
| Extraction protocol |
TF DBD sequence plus 15 a.a. N- and C-terminal flanking sequences were cloned into the pDEST15 expression vector, conferring an N-terminal GST tag. N-terminal GST-fusion TF DBD proteins were expressed using the PURExpress in vitro transcription/translation kit (NEB) following the manufacturer’s recommendations.
|
| Label |
Alexa488
|
| Label protocol |
Arrays were incubated with an Alexa-488-conjugated rabbit polyclonal anti-GST antibody (Invitrogen A-11131).
|
| |
|
| Hybridization protocol |
First, microarrays were double stranded by on-slide primer extension, as described in Berger et al. (Berger et al. Nat Biotechnol. 2006 (PMID 16998473)). The primer extension mixture contains Cy3-conjugated dUTP, to allow for quantification of the relative amount of double stranded DNA on each spot. Double stranded microarrays were pre-wetted in standard phosphate buffered saline (PBS) (10mM PO43-, 137 mM NaCl, 2.7 mMKCl, pH7.4) / 0.01% Triton-X 100 for 5 minutes. PBS / 2% (wt/vol) nonfat dried milk (Sigma) was applied to individual chambers of an eight-chamber gasket cover slip in a steel hybridization chamber (Agilent), and the assembled microarray was incubated for 1 hour at room temperature. Microarrays were then washed once with PBS / 0.1% (vol/vol) Tween-20 for 5 min and once with PBS / 0.01% (vol/vol) Triton X-100 for 2 min. Proteins were diluted in a 75-μl protein binding mixture containing PBS / 2% (wt/vol) milk, 51.3 ng/μl salmon testes DNA (Sigma), 0.2 μg/μl bovine serum albumin (New England Biolabs). Pre-incubated protein binding mixtures were applied to individual chambers of an eight-chamber gasket cover slip in a steel hybridization chamber (Agilent), and the assembled microarrays were incubated for 1 hour at room temperature. Microarrays were again washed once with PBS/ 0.5% (vol/vol) Tween-20 for 3 min, and then once with PBS/ 0.01% (vol/vol) Triton X-100 for 2 minutes. Alexa-488-conjugated rabbit polyclonal anti-GST antibody (Invitrogen) was diluted to 50μg/ml in PBS / 2% (wt/vol) milk and applied to individual chambers of an eight-chamber gasket cover slip in a steel hybridization chamber (Agilent), and the assembled microarrays were incubated for 30 minutes at room temperature. Microarrays were washed twice with PBS / 0.05% (vol/vol) Tween-20 for 3 min each, and once in PBS for 2 min. After each incubation step, microarrays and cover slips were disassembled in a staining dish filled with the first wash solution. All washes were performed in Coplin jars at room temperature on an orbital shaker at 125 rpm. Following each series of washes, microarrays were rinsed in PBS and slowly removed to ensure removal of detergent and uniform drying.
|
| Scan protocol |
PBMs were scanned in a GenePix 4400A microarray scanner. At least 3 scans were taken for each slide at different photomultiplier tube (PMT) gain settings. We refer to the set of array chambers for a single allelic series assayed on the same slide (up to 8 array chambers per slide) as an allelic replicate. For each allelic replicate, an optimal scan was selected, corresponding to the PMT gain with the lowest proportions of both over-saturated probes (over-saturated probes are defined here as having intensity levels of >215.5; the signal is completely saturated at 216 = 65,536 intensity levels) and under-saturated probes (defined as intensity levels < 25) across all chambers in the allelic replicate. Most often, this corresponded to PMT gains between 400 and 500. The same PMT gain value was typically used within an allelic series.
|
| Data processing |
Pre-processing procedures, including background subtraction, and spatial de-biasing, were applied to each array independently as previously described . Probes with foreground intensities less than background intensities were excluded. An empirical reference generated by pooling across Cy3 scans from 90 arrays was used for Cy3 normalization. Probes with log2 Cy3 observed-to-expected ratios greater than 1 or less than -1 were also excluded. Arrays were normalized within allelic replicates against a single anchor sample within the replicate, typically the reference allele. A log-scale additive normalization factor was estimated for each non-anchor sample using the trimmed mean of M-values approach originally proposed for RNA-seq normalization. Arrays were then normalized across allelic replicates using a single anchor allele present in all allelic replicates considered, typically the reference allele, with the exception of two allelic series (HOXD13 and SIX6) for which we used multiple anchor variant alleles (and averaged normalization factors across these multiple anchors). First, a reference sample value was calculated for the anchor allele using the log-scale mean of the anchor samples across the replicates. Then, to account for both shift and scale differences between allelic replicates, both a log-scale additive and a log-scale multiplicative factor were calculated for each replicate against the cross-replicate normalization sample. The log-scale additive factor for each replicate was estimated as the difference in the log-scale median probe intensity of the corresponding anchor sample and log-scale median probe intensity of the cross-replicate reference sample value. The log-scale multiplicative factor for each replicate was estimated as the median ratio of the rank-ordered and median-centered log-probe intensities between the corresponding anchor sample and the cross-replicate reference sample value. This is approximately equivalent to the slope of the log-scale QQ plot of the anchor and cross-replicate reference samples. Allelic replicates with low relative signal (as compared to other replicates) were deemed as lower-quality experiments and therefore filtered out at this step. This was assessed by comparing the upper-tail width (90th to 99th percentile) of the distribution of probe intensities for reference TF DBD samples in each allelic replicate. In most cases, at least two replicates were assayed for each member of an allelic series; for a small number of published data 1 reanalyzed for inclusion in this set, a single variant array was analyzed by comparison with replicates of the reference allele. If the tail width of a reference TF sample in an allelic replicate was less than 25% of the maximum tail width (or 50% of the maximum, if the array had been used previously and then stripped for re-use of the array in a new experiment) of the reference TF DBD samples across all replicates of the allelic series, the entire allelic replicate was excluded due to it being deemed as a lower quality replicate. A threshold of 50% of the maximum tail width was applied if the array had previously been used and then stripped for re-use of the array in a new experiment. Normalization across allelic replicates was repeated as above after removing the lower-quality experiments. After cross-replicate normalization, mixed-effects models were fit for each 8-mer aggregating across allelic replicates and probes to calculate 8-mer-level affinity scores. Summarization was performed for each 8-mer by first estimating probe-level means and variances across replicates using the limma smoothed variance estimator. Each probe-level estimate was further corrected for sequence position bias to account for differences in measured binding due to the distance of the 8-mer from the beginning or end of the probe sequence. Finally, 8-mer-level affinity scores were calculated using the two-step DerSimonian and Laird estimator across all probes containing the respective 8-mer sequences. All inference was performed at the 8-mer-level.
"affiniteEstimate," an 8-mer-level affinity score
|
| |
|
| Submission date |
May 31, 2023 |
| Last update date |
Dec 05, 2023 |
| Contact name |
Stephen Gisselbrecht |
| E-mail(s) |
[email protected]
|
| Phone |
857-540-2853
|
| Organization name |
Brigham and Women's Hospital
|
| Department |
Division of Genetics, Department of Medicine
|
| Lab |
Martha L. Bulyk
|
| Street address |
77 Avenue Louis Pasteur, Rm 468
|
| City |
Boston |
| State/province |
MA |
| ZIP/Postal code |
02115 |
| Country |
USA |
| |
|
| Platform ID |
GPL33450 |
| Series (1) |
| GSE233827 |
DNA binding analysis of rare variants in homeodomains reveals homeodomain specificity-determining residues |
|
| Supplementary file |
Size |
Download |
File type/resource |
| GSM7437478_151124_V14_199_Cy3_lp550pg100_6-8.gpr.gz |
3.3 Mb |
(ftp)(http) |
GPR |
| GSM7437478_151130_v14_201_Cy3_lp550pg100_6-8.gpr.gz |
3.3 Mb |
(ftp)(http) |
GPR |
| GSM7437478_151202_v14_199_HOXD13-REF_HOXD13-Q325R_HOXD13-I322L_HOXD13-R306W_HOXD13-S316C_HOXD13-Q325K_HOXD13-I297V_HOXD13-N298S_Alexa488_lp550pg100_6-8.gpr.gz |
3.3 Mb |
(ftp)(http) |
GPR |
| GSM7437478_151202_v14_201_HOXD13-REF_HOXD13-Q325R_HOXD13-I322L_HOXD13-R306W_HOXD13-S316C_HOXD13-Q325K_HOXD13-I297V_HOXD13-N298S_Alexa488_lp450pg100_6-8.gpr.gz |
3.0 Mb |
(ftp)(http) |
GPR |
| Processed data are available on Series record |
|
|
|
|
 |
