糖心TV

FAQ

1. How is ThermoPhyl different than other primer design programs like PrimerBlast?

A: ThermoPhyl fills a slightly different niche space than other programs as summarized in the following table:

Software	Database size	Remote/Local	FISH Assays (single oligo)	Traditional PCR Assays (two primers)	Quantitative PCR Assays (two primers and probe)	Through put	Summary output of sensitivity and specificity
Primer Blast	Large (Genbank)	R	N	Y	N	Low	N
Probe Check	Linked to public databases	R	Y	Y	N	High	N
Primique	Small to Large (user-defined)	R	N	Y	N	High w/ speed limitations	N
Thermo Phyl	Small to Large (user-defined)	L	Y	Y	Y	High	Y

2. What鈥檚 the best way to design candidate assays?

A: For traditional PCR, the most efficient way is to compile all (or a representative set) of your target sequences into a single fasta file, and simply feed this into BatchPrimer3. For qPCR, we have had good results with ABI鈥檚 PrimerExpress program, although each sequence must be input singly and the output compiled into a single file. We typically generate 50 candidate assays per target sequence, which can be defined in the primer design program.

3. My output from BatchPrimer3 is formatted strangely. How do I change it?

A: No need to worry. ThermoPhyl has an option to use the output from Batch Primer3 as is. Just select the appropriate option in the introductory screen.

4. If I design so many assays for the similar sequences in my target group, won鈥檛 I end up with a bunch of redundant assays?

A: Good question. Yes, you probably will, but not to worry, ThermoPhyl will detect these and filter them down to a unique list.

5. Can I test primers to target all the sequences in a fasta file?

A: Yes, just don't specify any sequences in your target list (i.e. save it as an empty file).

6. ThermoPhyl tells me I have some discrepancy between my target list and my sequence database and so can鈥檛 calculate properly the number of matches to targets vs non-targets.

A: Common causes of this problem include strange characters (such as &*,!%) in sequence names (in the 鈥渙utside_world.fas鈥�) file and/or the target list (鈥渢arget_list.txt鈥� file), non-unique identifiers (鈥楢Y100鈥� with file of sequence names which include AY100908, AY100909, etc.), and duplicate entries in the sequence database (鈥渙utside_world.fas鈥� file). Correcting these errors should resolve the problem. ThermoPhyl will still run despite these errors, but the sensitivity and specificity calculations will be adversely affected.

Common problems with sequence databases include duplicate entries, gibberish or inconsistent naming schemes, etc. Remember, 鈥淕arbage In, Garbage Out鈥�.

7. ThermoPhyl doesn鈥檛 find any matches in my dataset.

A: This can reflect the truth (although matches should always be found among your target sequences!), but more often is an artifact of either having the columns in the assay file (鈥渃andidate_assays.txt鈥�) in the wrong order (correct order is Forward, Reverse or Forward, Probe, Reverse). Another common problem is having the Reverse Primer sequences reverse-complemented. They should be listed as true Reverse Primers as you would order the oligos (i.e. written 5鈥� – 3鈥� for the opposite strand; this is format used by Batch Primer3 and Primer Express to output their assays). The script automatically reverse-complements them to search on the same strand as the Forward Primer. Finally, the primer file (鈥渃andidate_assays.txt鈥�) should have no special characters other than tab-delimitations and newline characters as in the example.

Finally, make sure you aren鈥檛 mixing 鈥楾鈥檚 and 鈥楿鈥檚 in your primers and your outside world file.

8. How long does ThermoPhyl take to run?

A: In tests on moderately fast WinXP machines (e.g. 2 GHz Pentium CPU w/ 3 Gb of RAM), we have tested 5,000 candidate qPCR assays against a database with 5,000 taxa, for 25 million comparisons in about 2.5 hrs. Most users will have many fewer comparisons than this and for most applications, ThermoPhyl produces output in seconds to minutes.

9. Why doesn鈥檛 ThermoPhyl search for degenerate bases/mismatches in my probe(s) and/or primers?

A: Two reasons:

First, because the premise of ThermoPhyl is to start with thermodynamically optimized assays, any introduction of degenerate bases is considered to be a compromise. For qPCR, ABI recommends against the use of degenerate primers or probes. Ideally, one should be able to find a sensitive and specific assay for a given target group without introducing degenerate bases.

Second, ThermoPhyl is designed to summarize large datasets and the output of searches with varying numbers of mismatches would overwhelm most users and defeat the purpose of the program. We always recommend comparing several of the best assays to your alignment file, and if absolutely necessary, introducing degeneracies at that point.

10. Can ThermoPhyl evaluate FISH probes as well?

A: Yes. Just select the option for a single oligo in the opening screen. The input file 鈥榗andidate_assays.txt鈥� should have just a single column.

11. What are the output files and how do I view them?

A: ThermoPhyl produces two output files. The first file (鈥榮orted_search_results.txt鈥�) is a summary table with a sorted tally of the number of target and non-target sequences matched by each candidate assay. The second file (鈥榬aw_search_results.txt鈥�) contains details of each match, including the amplicon length, the position on the DNA strand where each primer/probe matched, etc. These files can be viewed conveniently in a spreadsheet like MS Excel.

12. Is the 鈥楻_motif鈥� sequence printed out by ThermoPhyl ready to order as a Reverse Primer?

A: No, it needs to be reverse-complemented. ThermoPhyl prints it out this way to facilitate searching on the same strand by the user when comparing output.

13. Everything worked fine, now can I just order the first assay on the list?

A: For some applications, that may be OK, but in general we highly recommend looking at several of the highly ranked assays in a multiple alignment file (the ARB editor is really good for this) to confirm sensitivity and specificity, check for possible degeneracies, etc.