This form help page explains how to submit the Meva consultation form.
In the consultation form, a tooltip help is available if you go with mouse over the parameter names (except Netscape V.4), and a more detailed help if you click onto the parameter names.
The parameters shown here determine what data will be analysed.
Bibliographic Field. Choose the bibliographic field to be analysed. Important fields are f.i. Author, Affiliation, Title or MeSH Terms. You can find a complete listing of the fields and their meaning in the Field Help.
Meva shows all type 1 fields of all articles with their corresponding article number (PMID). If you select also a second field, all related type 2 fields of the articles will be shown as well. For huge data, leaving field 2 empty reduces the processing time and the net traffic load.
Search Filter. If you search for the bibliographic field MeSH codes, you should provide an asterisk (*) as filter for this bibliographic field for two reasons: First to exclude generic terms like Human, Animal etc. Second to limit your retrieval to MeSH major topics discussed in the articles. (MeSH major topics are marked with an asterisk inside the PubMed file and displayed accordingly in the Meva result.) This narrows the focus of your search and reduces the result size.
If you search for other fields, you may want to enter a string which should form part of the field. F.i. looking for Author fields you may want to restrict your Meva result to authors with the name 'Smith'. Provide Smith as a filter for Authors then.
If in doubt leave this parameter empty!
Besides, you can use the Boolean operators AND, OR and NOT to combine search expressions. In most cases you will not need this feature. If in doubt leave this parameter empty!
The operators are processed in a sequential order and do not adhere to any inherent precedence rules. Any precedence can only be forced by putting the appropriate subexpression into parentheses. An expression like a AND b OR a AND c equates thereby implicitly ((a AND b) OR a) AND c. Searching for Colchicine AND blood OR pharmacokinetics and for Colchicine AND (blood OR pharmacokinetics) thereby usually doesn't find the same. Two operands without an operator between are AND combined. An expression like Colchicine blood equals thereby Colchicine AND blood.
At the moment, the operators work only at field level and not at article level, so the operands refer to the same field! If you look for Colchicine AND blood, Meva will find Colchicine/blood, Colchicine/blood/pharmacokinetics and also Colchicine/*analogs & derivatives/blood, but not fields containing only Colchicine, only Lysine/*analogs & derivatives/blood/pharmacokinetics or only Colchicine/pharmacokinetics - even if they all occur in the same article! This can be better done in PubMed itself using its Boolean operators, f.i. Colchicine[mh] AND blood[mh]
To find complete strings containing several words, put the string into double quotes. Searching for "Cell Line" will find thereby only fields containing the whole string, but not fields containing only Cell, only Line or only Cell Proliferation Line.
To refine your search, you can set additional search parameters:
If the checkbox Ignore Case is active, searching for age or Age will find the same.
Deactivate the checkbox Whole Words Only if you are looking for substrings. The blank is regarded as word border. If activated, searching for age will find only whole words like Age, age and AGE (depending from case option), but not Aged or Heritage.
Ex. 1: You want to look for all words beginning with Age (Age, Aged or Aging, but not Dosage or Heritage). Leave the case option as well as the whole words only option deactivated. This is the default.
Ex. 2: You want to look for words with hydro inside no matter the case, f.i. Hydrofolate, but also Dehydrogenase. Activate the case option and leave the whole words only option empty.
Ex. 3: You want to perform an exact search for Age. Leave the case checkbox empty and activate the whole words only option.
Ex. 4: You want to find Age as well as age, but not Aged, aged etc. Activate both checkboxes.
For authors, you can activate the appropriate radio button to restrict the analysis to First Authors or Last Authors.
Minimal Frequency. You can restrict the result onto fields with minimal frequencies. If you enter 3 f.i., only fields will be shown which occur at least three times in the PubMed result. Recommended to reduce the net traffic!
The parameters shown here determine, what will be printed out and how.
Diagrams. Please select if histograms of the most frequent fields as well as coincidence tables shall be returned.
Determine the count of fields listed in the diagrams entering a value into the Top input field. If you wish to see only three fields in the diagrams, enter a 3 here. If this value exceeds the overall count of fields found, it will be reduced automatically by Meva. The maximum count allowed is 50.
Adjust the optical size of the histograms providing a value into the Scaling input field. In most cases, it should be left on its default value auto. In rare cases however when you want to compare histograms of several Meva results optically, it can be helpful to set the scaling factor onto an absolute value.
Details. Select if you wish to retrieve the details of the database.
Furthermore choose if you want the details sorted alphabetically or by frequency and in which sort order, ascending or descending.
The checkbox Always show PMID forces Meva to print the PMID's even if you leave field 2 empty. By default this is deactivated in order to retrieve a very compressed detail listing for a single field only.
MeSH Codes. Choose if you want MeSH terms accompanied with their codes inside the several sections. Remember not all terms own a code so don't be surprised to find some terms without a code, especially generic terms like Human or Animal. Activate that option only if you need it since it slows down the processing. This option is not needed for the MeSH tree and in most cases you won't need it.
MeSH Tree. If you have chosen MeSH terms for field 1 or 2, you can opt for a MeSH tree for print out: all MeSH terms according to their codes and along with their frequencies will be shown. Adjust until to which Depth the tree shall be printed out. A depth of 1 f.i. shows only the main categories.
Since PubMed delivers only the string of a MeSH term, but not its code and due to the fact that a term can have several codes - Orientation f.i. can be found in F01.058.577 as well as in F02.830.606 - Meva shows all corresponding codes. But now the problem arises, which code should be charged with the frequency of the term. If f.i. Orientation could be found 9 times, which code shall bear the 9: F01.058.577 or F02.830.606? Meva tries to solve that problem with several strategies:
Branch fit. The user can select a preferred MeSH Branch - a main category - to be charged with the frequencies. If there's no luck with this, maybe because of an unlucky user choice, maybe since many codes are in the same branch, Meva tries another strategy:
Deepest fit. Here now Meva chooses that code which bears the deepest code in the tree hierarchy to increase the specificity of the result. If this fails as well, Meva switches onto the last strategy:
First fit. Here the first code to be found will be charged with its frequency. Obviously this is the case in the example above.
Link Restrictor. Meva generates search links to PubMed inside its result - provided you have entered a valid PubMed search string into the link restrictor input field. This value will be combined with the field values of Mevas result to form a search link. For instance, if you have provided Tetrodotoxin[mh] as a link restrictor and you click onto the author Smith TK inside Mevas result, you will trigger a new search in PubMed for Smith TK[au] AND Tetrodotoxin[mh].
What is it good for? Since Meva doesn't "know" which search expression you have used in PubMed to retrieve your original result (primary solution set), you can pass the information this way to Meva. Thus it is ensured that the solution set of a this new search (secondary solution set) does not exceed the primary solution set - in other words, you can refine your original solution stepwise.
The link restrictor string must comply with PubMeds search syntax, otherwise you won't get a sensible result. It is advisable to search in PubMed always for specific fields instead of for text words in order to get a more specific result. Simply put the field tag in square brackets behind your search expression. For instance, to look for MeSH terms, Authors or Publication Dates, use Tetrodotoxin[mh] instead of Tetrodotoxin, Smith TK[au] instead of Smith TK and 2002[dp] instead of 2002. Don't use PubMed references like #1 as link restrictor since they are only valid for a short period of time.
In case you haven't provided a link restrictor, Meva won't produce links (except for PMID and correlation tables) to protect PubMed from unspecific global searches: a search for Smith f.i. in PubMed is of little use and loads PubMeds database unnecessary.
Some examples will demonstrate the usage of the link restrictor in more detail:
Ex. 1a - Reduction of original solution set: You have searched initially for Tetrodotoxin[mh] in PubMed in order to retrieve an input file for Meva. Besides, you have chosen Author as bibliographic field to analyze inside the Meva consultation form. Now if you click onto one of the authors in Mevas result, you want PubMed to find only articles related to this author and the MeSH Term Tetrodotoxin. In this case, you provide the original PubMed search expression Tetrodotoxin[mh] as link restrictor. Thereby you ensure that PubMed will retrieve only articles which are part of the original solution set.
Hint: This constellation is useful if you won't exceed the original solution set.
The picture below shows the counts of the solution sets (retrieval date: 06/2003). The left circle illustrates the primary solution set Tetrodotoxin[mh], the hatched area the secondary solution set Smith TK[au] AND Tetrodotoxin[mh], resulting from clicking onto Smith TK within the Meva result. As you can see, 18 articles would be found:
Ex. 1b - Further reduction of original solution set: As in the previous example, you have searched in PubMed for Tetrodotoxin[mh] and selected Author as bibliographic field in the Meva consultation form. Now if you click onto one of the Meva result author links, you want PubMed to find the same like in ex. 1a, but limited onto the publication year 2002. In this case, you extend the link restrictor to Tetrodotoxin[mh] 2002[dp] (or, which is the same, Tetrodotoxin[mh] AND 2002[dp]). Of course you can choose another restrictor as well.
Hint: This constellation is useful if you wish to retrieve an ever smaller secondary solution set.
The upper left circle illustrates the primary solution set Tetrodotoxin[mh], the hatched area the secondary solution set Smith TK[au] AND Tetrodotoxin[mh] AND 2002[dp], resulting from clicking onto Smith TK within the Meva result. As you can see, 2 articles would be found:
Ex. 2 - Partial extension of original solution set: As in the previous example, you have searched in PubMed for Tetrodotoxin[mh] and selected Author as field in the Meva consultation form. Now if you click onto one of the Meva result author links, you want PubMed to find all articles (not only Tetrodotoxin related ones) from this author of the year 2002. In this case, you provide only 2002[dp] as link restrictor. The link restrictor in this case is only needed the let Meva know to generate links. Of course you can choose another restrictor as well.
Hint: This constellation is useful if you wish to know what else an author has written on.
The left circle illustrates the primary solution set Tetrodotoxin[mh], the hatched area the secondary solution set Smith TK[au] AND 2002[dp], resulting from clicking onto Smith TK within the Meva result. As you can see, 15 articles would be found:
Result format. Select if you like to retrieve the result as HTML or as plain text.
With HTML, you can have diagrams, links and extended formatting whereas with plain text you can import it into a database. You can take a snapshot from the browser window (Alt-Print) below MS Windows and then paste into every application (Word, PowerPoint etc.) for presentations.
Hint: Many browsers allow to display a page without annoying frames in full screen pressing F11.
The text file comes in TSV format, it contains Details or MeSH section data separated by tabs. The 1st line contains the field names. Many databases or chart programmes allow to interpret the 1st line of a text data file (SDF, TSV, CSV etc.) as field names. If not - you can simply remove the 1st line.
Comment. You can provide a user comment here, which will be embedded into the result. This facilitates to identify Meva search results later on.
There are two good reasons to use that field:
Since the comment will be included into the result title and since many browsers as f.i. Internet Explorer use the title of an HTML file as default file name for saving the file, you will get an evident file name for your saved Meva results.
Besides, the comment appears in the headline of every page printed out. This enlightens keeping pages from the same search together.
If left empty, the link restrictor will be filled in.
File Name. Enter the complete path to the PubMed result file, f.i. f:\in\ms20020617.txt (Windows) or /home/user/ms20020617.txt (Unix).
Consult Meva. This button sends the form data to Meva for analysis.
Clear. This button clears all form data.
PubMed Restrictions. Please adhere to NCBI's policy (!): "Do not overload NCBI's systems. The Entrez search and retrieval system was not designed for downloading large quantities of data. Users intending to send numerous queries and/or retrieve large numbers of records from Entrez should comply with the following: Run retrieval scripts between 7 PM and 5 AM Eastern Time weekdays or any time on weekends. Make no more than one request every 2 seconds." That way you will help to keep that service publicly accessible and free of charge!
Meva Restrictions. To protect the network, Meva denies to work if the input data are very large (this limit can be changed by the administrator). In this case you can compress the data using MePrep, the Meva Preprocessor, locally onto your computer before sending them to Meva. The compression reduces your data usually onto 1 - 20% of the original file size. F.i. if Meva is restricted to accept not more than 5MB from the user, you can send virtual 25 - 500MB to Meva anyway. In general, if you have large data to analyse or repeated consultations with same data but different parameters, it is recommended to use MePrep since it will decrease the net traffic load and increase the processing speed of client and server.
Despite the fact of the restricted input size, Meva has output limitations as well to protect web clients from dealing with megs of data. The output limitations relate only to HTML mode, not to text mode result.
