PLACE

A2:
As far as I know, not only we, but also any other cis-element databases similar to ours, can NOT guarantee that. We survey many journals by ourselves, and read many papers. You need to realize that, since human being, not the computer, survey publications and read them, there would be some errors, and some motifs may be inadvertently missed. No authors have ever deposited the motif sequence to us prior to or after the publication of their papers. Many journal publishers require payment for downloading pdf files, thus it is getting more difficult for us to obtain a copy of important papers soon after publication.

A3:
Yes, there are many. TRANSFAC is the oldest of its kind, and includes animal, yeast, plant motifs and transcription factors data. Some web sites providing the service similar to ours have their own database, at least they claims so, and others use PLACE and TRANSFAC database. You can obtain information by Internet search.

A4:
In order to compare the occurrence of some cis-elements between two populations of promoters, and to make sure that the difference or preference observed are really significant, you will need to carry out a number of statistical analyses of the results of PLACE analyses.
Let me refer you to the following interesting publication.
"Interdependency of brasssinosteroid and auxin signaling in Arabidopsis" by Nemhauser, Mockler, Chory, which appeared in PLoS Biology September 2004/Volume 2/ Issue 9/e258. This can be found at www.plosbiology.org.

A5:
We can not provide that kind of service now, mainly due to the lack of manpower.

A6:
These two entries are:
MYCATERD1 (S000413) CATGTG
MYCATRD22 (S000174) CACATG
SignalScan program search the query promoter sequence for the presence of motif sequences in the SENSE and ANTISENSE directions.
If CACATG (MYCATRD22) is found in the sense strand of the query, the result, in the case of table format, is shown as, e.g., 219 (+) CACATG. If CATGTG (MYCATERD1) is found in the antisense strand of the query, it is shown as, e.g., 219 (-) CATGTG..So, two "different" motifs are shown at the same site. But, if you look carefully at the motif sequences, you will notice that these two motifs are "complementary" to each other.
MYCATRD22 (S000174) 5'-CACATG-3'
Antisense of this sequence is 3'-GTGTAC-5'
If you read the bottom sequence form the tail, it is 5'-CATGTG-3'. It is the same sequence as MYCATERD1 (S000413) CATGTG. These entries are based on the original reports as they appeared in Journals. MYCATRD22 and MYCATERD1, both, have been reported by the same research group in two different genes (RD22 and ERD1) as the MYC-like transcription factor binding sites, as I recollect, from Arabidopsis thaliana (AT). If we are sure that these two motifs are exactly identical in function, then we will combine these two entries.

A7:
Please compare these two systems by yourself.

A11:
We compile the motifs or the consensus motifs reported in the papers. Thus, some motifs are registered in the PLACE database as ACGTGC, or GTGCAC, etc. Some papers report the consensus sequence of the motifs, e.g., ACWTGC, etc., where W=A/T. Some consensus sequence look like CWWWWWWWG, or ANNNTGC. In the latter case, Signal Scan program allow any nucleotides between A and T. So yes, the Signal Scan result report shows only 100% identity, but that allows some flexibility if the registered sequences contains W or Y or R or K, etc.

A12:
We do not create that. I do not think we have sufficient number of motifs to create reliable matrix for plant cis-elements. Our policy* is to collect motifs which have experimental evidence. But since the genome sequences of Arabidopsis and rice, and also microarray data became available, in silico search of cis-regulatory DNA elements become feasible, and we will soon get a flood of such motifs. Then, we may be able to create matrix. (*Note that we have changed this policy in late 2005. See the release notes.)

A14:
Yes, I noticed that there are so many similar or identical motifs in this database. In some entries, probably I can simply summing up these similar motifs and replace them by "super-consensus motifs," or something like that. Many MYB-related motifs may be the case.
In other cases, the motifs reported in several publications are slightly different so that it is difficult for us to choose which motif should be entered in to the database. But, can we exclude a possibility that the slightly different motif sequence might be due to species-specific binding site? So we entered all these motifs in to the database as they were reported.
In other cases, there is evolution (!) of the motif sequences in publications. Usually, longer sequences were reported in earlier publications, and we found shorter sequences in later publications.
If you have any suggestion, please let me know.