Project Documentation & Protocols: UniformMu Mus2Use Transposon Resource

Release notes - May, 2009:

Our first set of UniformMu germinal insertions mapped in the maize genome is now available at MaizeGDB, and the corresponding consensus W22 insertion flanking sequences are deposited in Genbank. Individual seed stocks containing these insertions are now available for distribution through the Stock Center.

In MaizeGDB, the mapped insertion sites are displayed in the genome browser and each insertion is linked to seed stocks available through the Maize Stock Center. Each insertion site has a unique identifier (e.g. mu1011038) that can be used as a keyword to search MaizeGDB.

The links returned conform to the familiar locus/variation/stock theme used by MaizeGDB. Thus, each insertion site has a locus page (e.g. mu1011038) which includes a link to a single variation page (e.g. mu1011038::Mu) describing that insertion. The variation page in turn includes a link to a page describing the corresponding seed stock (e.g. UFMu-00392). The stock display contains a link in the right hand panel for requesting seed directly from the Stock Center.

Note that a single stock may be linked to multiple insertions depending on how many insertions have been mapped in that line. The stock page lists all insertions currently assigned to that line. The number of unique germinal insertions per line ranges from 1 to 50.

Importantly, this means that phenotypes associated with the seed stock will not necessarily be linked to the insertion of interest.

To search Mu insertion sites by BLAST:

A popular method for searching the insertions will be by BLAST. At present, this can be done in two ways:

One way is by using NCBI's standard BLAST service. To search the UniformMu consensus flanking sequences, select the standard nt database and enter "UniformMu" as an Entrez keyword in the NCBI BLAST form. The Entrez keyword will restrict the search to UniformMu sequences.

The most direct route to finding seed for an insertion identified by BLAST is to enter the unique identifier (e.g. mu1011038) of the insertion site in the MaizeGDB search window and click on the link to its "stock record". Then click the "Order this Stock" link in the right hand side panel.

The second way is to use our BLAST site at uniformmu.uf-genome.org. This database consists of 1kb flanking sequences extracted from the B73 genome. We have found that these 1 kb segments centered on the insertion site are convenient for annotation purposes as well as for designing PCR primers for confirmation and genotyping of Mu inserts. (Just to be confusing, we used a slightly different format for the insertion identifiers in our current BLAST database, e.g. MI_1011038 in place of mu1011038, but MaizeGDB should see the MI_ and mu prefix formats as synonymous. We'll conform to the MaizeGDB format in future releases.)

Downloading the flanking sequences from Genbank:

The complete set of consensus sequences can be retrieved from NCBI using the Entrez keyword "UniformMu". The Genbank entries are gapped consensus sequences derived by assembly of the anchored 454-reads obtained from the UniformMu plants (I.e. W22 genome). In cases where both sides of the Mu insertion have been sequenced (a majority of cases) a gap of 100 N's has been inserted to indicate the position of the transposon. Hence, the gap will be flanked by the expected 9-bp direct repeat created by Mu. Entries that do not include a gap have sequence from only one side of the transposon, and in those cases, the Mu insertion is located at the 5'-end of the Genbank sequence. I.e. these sequences are oriented outward from the Mu TIR. The consensus sequences vary in length, a few insertions that are listed in MaizeGDB and searchable at uniformmu.uf-genome.org are not included in Genbank because they fall below NCBI's minimum length requirements. These will be submitted to Genbank as more sequence data is obtained.

PCR confirmation and detection of Mu insertions:

We have extensive experience in designing successful primers for PCR genotyping of Mu insertions in the maize genome (Primer design notes - under construction) achieving about 85% average success rate with the first pass primer design. In our experience, primer design issues account for the large majority of confirmation failures (Settles et al. 2008). Hence, if the initial PCR tests fail, redesigning of gene specific primers flanking the insertion usually solves the problem.

Known issues:

Finally, there is a known issue affecting this first dataset caused by cross-contamination of our multiplex 454-sequencing adaptor oligos at the manufacturer. It's a long story but after receiving 2 consecutive batches of cross-contaminated adaptor oligos we finally convinced our oligo supplier that people doing NexGen sequencing are going to notice and care about low level cross-contamination in batch oligo orders. Working with the manufacturer, we were able to suggest modifications to their purification protocols that eventually eliminated the problem. We've filtered out the noise as much as possible, and while we have done extensive PCR validation of the axis assignments in this dataset, there is still a small chance that some insertions will be assigned to the wrong seed stock.

In the event you are unable to confirm an insertion by PCR, please contact us (drm@ufl.edu) and we will assist you in identifying the correct seed stock.

Future data releases:

The next data release is planned for July, 2009, which we anticipate will include ~12,000 unique insertions, mapped in a set of 2451 F3 UniformMu lines. The total number of unique insertions in the collection will grow steadily over the next two years. The pace is determined primarily by the rate of production of high quality, Mu-inactive F3 seed stocks. We believe that a total of 100,000 sequence-indexed germinal insertions is an attainable goal for this project.

Return to Documentation Index | Return to Homepage