Resources - Questions and Answers

Note: Names and other identifying information have been removed from any correspondance listed on this page.

"I ordered 2 BACs from you from the RPCI23 mouse BAC library; clone numbers RP23-201B12 and RP23-301F3. These 2 clones are supposed to contain genomic fragments of a mouse gene which is located on chromosome 1. However, I was unable to PCR-amplify any sequence using the gene-specific primers. And when I subcloned sequences from the RP23-265D16 clone, I got genomic sequences belonging to mouse chromosome 10.

Where is the mistake coming from? and what should I do? Maybe I ordered the wrong clone, so here is the genbank document corresponding to one of the clone (RP23-301F3) I wanted"

We have looked at your request and tried to find a suggestion for the mismatch between the clone ordered and the sequence it contained. We looked at the history of the various replicates of the arrayed BAC library. In the original 384-well dish (#301) obtained after picking colonies, our records indicate well F3 had clear medium, meaning no growth of bacteria. After replicating the library for the first time, there was also no visual growth. Only after a second replication cycle, the sterility of the well was gone and the well was turbid, probably due to cross-contamination from cells derived from one of the neighboring wells. I am sure that your are aware that handling of thousands of microtiter dishes can not be done under conditions of absolute sterility. The remaining question is what the true identity of the clone in the database is. I presume that the clone in the database came from a copy of the library were well "F3" was contaminated and possibly heterogeneous. There is no way for us to know the true identity of the clone in the "master" copy of the library - without contacting the laboratory which originally screened the library and/or sequenced the clone, and probably not without additional library screening. The clone exists in our master-copy of the library, but likely at a different address (well position) in the library. In this particular case, we know from our record that the clone must have come from a different well because our original master copy had no growing cells as registered soon after library generation. Hence, the well might have become populated (in a replicate of the librray array) through well-to-well contamination. In addition, clones can make "virtual moves" within the arrayed library as a consequence of typographical errors when the database record were established or when the original clone name was incorrectly copied anywhere along the process of handling the clone prior to sequencing.

We just do our best to deliver what is most likely identical to your requested clone and at the lowest possible cost - in most cases the requested clones agree with the expectation (far over 90% of the time).

2. How can I read the results from the hybridization of the filters I ordered?

General instructions on how to read hybridization signals from our filters are available in Microsoft Word format from our download page.

4. I am interested to work with genes of the Pseudoautosomal region of cattle and I would be grateful to you if you could send me the BAC clones for the genes in the PAR region of Cattle.

We cannot do this for you, but we can provide you the clone if you give us the coordinate in the library (for instance RP11-15G11 or CH201-14H1). Here's a page with popular mapping resources that you can use to obtain the clones of your interest: http://bacpacresources.org/mapping_res.htm since cattle genomes are not at advanced sequencing stages, you will probably have to rely on gene expression data, which may not be too helpful in deciphering genome mapping data; In this case, try to go to Vancouver genome sciences center (http://www.bcgsc.bc.ca/) where they have a java program called ice (internet contig explorer) which allows you to browse fingerprinting data of genomes in sequencing pipelines. They have a bovine database in that program which may help you in finding the clone you want provided that you have the appropriate markers for your gene of interest. In last resort, if you cannot find the clone for your gene of interest in the databases, you may want to order from us a filter set for the genome of your choice and do screening of these filters using an appropriate probe for your gene, in this case you will end up with about 10 clones (if the library is 10X redundant) that you can order from us.

5. I would like to order a BAC clone I have seen in a publication. The name is RP11-469H8. I don't see this clone listed on your web site. Can you let me know if you have it and how I can order it?

Answer: We have nearly 20 million distinct clones in our freezers at BACPAC Resources, CHORI, Oakland. All these clones can be ordered by following our ordering instructions http://bacpacresources.org/ordering_information.htm. We are not generating or maintaining data on most clones and therefore, we do not separately list each clone. An exception is a very small subset of clones isolated and characterized from the original libraries. If the clone is part of this small subset of mapped or otherwise characterized clones, then you can find this on specific web pages on our site (see: http://bacpacresources.org/pmapped-clones.htm). All other clones are maintained as part of microtiter dish-arrayed libraries. Most likely therefore, you will not find the clone as part of a characterized subset. This does NOT mean that we cannot provide the clone. To find out if we have the requested clone, you can follow any of two procedures described below:

First, you can dissect the name of the clone into it's components. For instance RP11-469H8 indicates that the clone is part of the RPCI-11 BAC library (abbreviated as RP11, see the NCBI nomenclature rules: ). It is in 384-well microtiter dish #469, Row "H" and column "8". You can check the properties of this library by inspecting our library list http://bacpacresources.org/libraries.php. You will find there a specific hot link to the RPCI-11 BAC library page: http://bacpacresources.org/hmale11.htm. This library was prepared in our laboratory and is maintained in 1,440 microtiter dishes numbered from 1 through 1440. The dish #469 is a valid number in this range and is part of library segment "2" prepared by cloning genomic DNA partially digested with EcoRI into BAC vector pBACe3.6. Hence, we most likely have this clone although we cannot guarantee that the date in the publication or in public databases indicates the correct clone. There are a significant number of incorrect entries in databases due to data and sample tracking problems. Hence if you order this clone from us, you will obtain viable bacterial cells in a bacterial stab culture to be confirmed in your laboratory and lacking a guaranteed identity. The odds of an erroneous clone/data association is about 10-20%, so please be aware and, please, check by PCR or another technique.

A second alternative is to use the NCBI Clone Registry (for BACs & PACs, not for cDNA clones): http://www.ncbi.nlm.nih.gov/genome/clone/index.html. Enter the clone name in the registry and you will obtain an answer if the clone has associated data in the NCBI database. If the person publishing the clone has not entered the clone into the database, you will obviously be out of luck. Following hot links on the positive NCBI reply, you will arrive at a page which summarizes all linked information. This page also includes distributor information. In this particular example, you will return to our contact information.

6. I have a human BAC RP11-96I16 of chr 18 for the n-cadherin gene and am trying to find the sequence of this clone. I went to the ucsc site but it does not list the BAC clone. Where can I get the sequence in fasta format for this clone?

You are making the same assumption many people make. It is wrong to assume that all clones found in mapping databases and map browsers have been sequenced. Most clones have NOT been sequenced. Here is why:

All BAC libraries are generated by random cloning of large DNA fragments. Any random process has a finite chance that some genomic fragments are not present. The odds of not having a complete genome representation increases with small libraries. Therefore, most libraries are at least 10-fold redundant and the human RPCI-11 BAC library is more than 30 fold redundant. This resulted in a minimal number of gaps in the genome represented in the library. It also allowed the selection of a set of mapped clones with somewhat minimal overlaps. The minimal overlaps are important for the economy of the human genome sequencing: shotgun sequence libraries were prepared from a minimal set of BAC clones. Hence, while the RPCI-11 BAC library is >30-fold redundant, less than 1.5-fold redundancy of the BAC clones have been subjected to shotgun-sequencing. About 10-fold redundancy of BACs have been mapped either by fingerprinting or virtually mapped by BAC-end sequencing.

How does one find out if the clone has been sequenced?

There is no answer which is suitable for all the sequencing projects and species. However, for the human genome and several additional genomes, you should first go to the NCBI CloneDB. Type the name of the clone and find out all mapping and sequencing information linked to the clone. If you don't find any information, then there are two options: you had a typing error in the clone name or there simply is no information available for the clone. To avoid typing errors, make certain that the name of the clone agrees with the NCBI clone nomenclature. Please, realize that many sequencing centers are not maintaining the clone names according to the nomenclature rules and you may have to convert the clone name into the NCBI-recommended name. In the example for the RP11-96I16 clone, you will find the following registry information (follow this link). If the clone registry indicates an accession number for a high-quality or draft assembly of the BAC sequence, then you have your answer. However, for most clones the only information displayed indicates that the BAC-insert ends have been sequenced. This is enough for map browser programs to map the BAC with great precision.

If the clone has NOT been sequenced, then you can nevertheless retrieve the equivalent sequence from the human genome. The "equivalent" sequence is derived from the genomic coordinates corresponding with the cloned fragment. This sequences is, however, derived from a different haplotype and from a different set of clones & libraries. Here is how you can retrieve the sequence. Go to the UCSC Human Genome Browser (http://genome.ucsc.edu/cgi-bin/hgGateway) and search for the clone name (in the correct NCBI nomenclature) by pasting the name into the "Position" window, replacing whatever text was already there. In the example for the RP11-96I16 BAC, you will find that a map is displayed for 148,874 bp (the approximate size of this BAC with the data based on the Dec. 2013 (GRCh38/hg38 Assembly). This BAC has not been sequenced but nevertheless the equivalent size is known with single base pair resolution. This is not the real size of the BAC but the size of the corresponding sequence from another haplotype. Hence, it is likely that the sequence is different with on average one SNP mutation per 800 bp and perhaps also insertion or deletion events. You can do many things with the displayed map. You can configure the map to display the BAC better (i.e. activate the BAC-End Pairs option to "Full" and then use the "Refresh" option on the web page - not the internet browser). You can also retrieve the equivalent DNA sequence, by activating the "DNA" button at the top of the map browser. Success!

7 Can you indicate where I can find the sequence information for clones I recently ordered? [phone message, August 31, 2005]

I found your order for clone RP22-481B2, RP22-535B14, etc. I am not sure how you obtained the names of these clones. The name "RP22" indicates the name of the BAC library (see our list of libraries), which is derived from the 129S6/SvEvTac strain, which is isogenic with some of the popular mouse ES cell lines. This is usually why people are interested in "129" BACs: to create targeting constructs from isogenic DNA. Unfortunately, "RP22" is not one of the BAC libraries used for the mouse genome project, Very few clones from this 200,000 clone library have, therefore, been characterized or sequenced. We are not the 'gate keepers' for the data generated from our BAC clones. The best way to find out if there is any data related to a particular clones is to search NCBI CloneDB. The registry is not complete, but is the best first approach. If the clone is not listed in the clone registry, then try searching in Genbank with a text search for the correct clone name. The correct clone name should follow the recommended NCBI clone nomenclature. The chance that you will find information related to these clones is, however, very, very small. As I mentioned, I have no idea how you found these clone names. Perhaps you found these in the literature and perhaps these were sequenced or mapped by somebody else? Please let me know if this information was useful. Thanks.

8 We received one wrong clone. Please send us the correct clone as part of your warranty service.

You order a clone and once the clone arrives, it appears that it does not contain the sequence attributed by the public databases to this clone. The first reaction is a logical one: "you paid for it and should get what you expect". Here is why this is not the case. We cannot provide a warranty that the databases are always correct. In fact, an error rate of around 10% should be expected: good data associated with the wrong clone name. As you will certainly understand, we cannot become an insurance company providing warranty services related to experimental data generated elsewhere.

Most of the clones were generated in our laboratory. This process involves streaking freshly transformed cells (prior to cell duplication) on large square dishes with LB agar and the required antibiotics. Colonies formed after an overnight incubation and these were arrayed into separate wells of 384 well dishes by a colony picking robot. All dishes (also known as "plates") are labeled starting with plate #1 and going in a continuous plate numbering to the final plate number in the library (any number below 3,000). This is the stage when clones also get there ad-hoc names. The naming convention is based on the NCBI clone nomenclature recommendations. This means, for instance, that a clone from the RP11 library residing in plate 128 at the intersection of row "D" and column "15" will be labeled: "RP11-128D15".

When you order a clone from us, we do not check public databases to find out which sequences have been reported for a particular clone, and do not determine if the sequences in the clone are consistent with the database. It is not economically feasible for us to check if the databases indicated the correct clone. We will simply find the requested clone in the freezer, then streak it to single colonies on a petri dish with LB agar and the appropriate antibiotic. We then take a single colony to inoculate an agar-stab culture (standard microbiology) with the appropriate antibiotic. Hence, the only thing we know about the clone being mailed, is that it very likely was correctly retrieved and that it is still viable (a fresh colony was obtained).

In nearly all cases when a clone fails to meet the expectation (in about 10% of the clone deliveries), the database was incorrect. In summary, we provide no warranty service. However, we will - at your insistence and as a courtesy - retrieve the same clone once more and ship it to you free of clone retrieval charges (except that we ask you to pay for the mailing costs). If you would like to receive a alternative clone for the same gene(s), then a new order will have to be placed online.

9. How do I find the name of the vector used for a particular BAC clone?

Each BAC clone is part of a BAC library. We have a specific web page for each library which describes the common properties of all clones for a particular library. How to know the library name? The library name is embedded in the clone name according to the NCBI Clone Nomenclature recommendations.

In this case, "RP23" is the NCBI library name and the full name of the library is "RPCI-23". The NCBI also has a library browser with all "approved" library names. Our own web site also contains a library browser with extensive detailed information on all BAC libraries we distribute clones for. The default version of this browser organizes the libraries according to a phylogenetic tree of species. Alternatively, the libraries can be viewed ordered by name as derived from the abbreviated name of the originating institute. Most of these libraries have been prepared in our own laboratory (CHORI or RPCI). The specific page for the RP23 mouse BAC library contains the name of the vector: pBACe3.6 and a link to the sequence and vector graphics.