Further clarifying "Expression Pattern"
Not to be flagged as Expression Pattern:
"A fusion of 6 kb of sequences upstream of the cog-1 start codon to gfp shows expression in the sites previously reported to express cog-1, namely vulval cells and head neurons, including ASER; We analyzed a dataset of 52 genes that based on reporter gene analysis are expressed in ASE (ETCHBERGER et al.2007)(Supplementary Table 1)." ----> send to Sequence Feature
In this example the expression patterns were determined previously and the authors do not indicate any new patterns of expression.
Wen does "...not curate controls since these data should have been published before and entered to WormBase. Authors usually list the original publication of these controls or wildtype patterns so it is clear that those are not original data. If in some cases you are not sure and feel that the data may have not been published before, you can check WormBase."
FROM WEN (some typos corrected): For firstpass, there is an easy way to estimate whether a paper contains Expr_pattern data, which is the purpose of the paper. The author must be studying one or multiple genes, and looking for the time and tissue distribution of these genes in real life. There are many cases that GFP was used but people were not really studying expression pattern. These cases are not Expr_pattern: 1. Expression in mutant background or under chemical, heat or cold treatment. They want to know how will the expression level or pattern change in different cases. These are Gene_regulation. In this case, authors already know how the expression is like in wild type background. 2. A GFP strain was used to label a cell so that they can see how the cell develop or migrate, or simply to find out the location of the cell. These are not Expr_pattern. If they used a integrated strain instead of just did the plasmid injections, the data are curated as tissue specific marker transgenes. 3. The paper is about developing a new technique, so they used a previously published GFP strain to see how good their technique is comparing with previous technique. They are interested in how good is the technique, instead of where the gene is expressed, so they only try genes that were studied and published before. These are not Expr_pattern. Maybe "no curatable"? :-) 4. They repeat experiments that where published before, and they found exactly the same results. Sometimes author have this at the beginning of Results section as the start of their story. These are not Expr_pattern either, unless authors indicate that they find something new. 5. They express gene B and GFP under a different promoter and want to see how will that change the phenotype, for example, rescue a mutant. In this case, the pattern they observed are not the expression pattern of gene B itself. This is gene function assay, not Expr_pattern. (possibly also gene-regulation, cis_regulation, overexpression, and site_of_action) 6. They already know where a gene is expressed, but they want to know which part of the promoter is the most important, so they made a series of GFP constructs with different promoter length to see which one display the right pattern. This is not Expr_pattern. This is cis-regulation, or Sequence Feature. There might be other cases. Once you know what is the goal of Expr_pattern class, it should be easy to tell from the nature of the paper.