QUALITY v1.1.6, 2013-09-12 Description ----------- The program QUALITY is a variant of the minimum Chi-squared (MC) method described by Taswell [2] for limiting dilution assays, and for which he has demonstrated by simulation desirable properties of minimum variance (i.e., high precision) and minimum bias. Our method modifies the MC method to allow the user to specify the probabilities of a false negative and false positive PCR. The program, and the methods used, are described in detail in Rodrigo et al. [1]. On line info and updated versions are available at 1. Rodrigo, A.G., P.C. Goracke, K. Rowhanian, and J.I. Mullins. 1997. Quantitation of target molecules from PCR-based limiting dilution assays. AIDS Res. Human Retrovir. 13:737-742. 2. Taswell, C. 1981. Limiting dilution assays for the determination of immunocompetent cell frequencies. J. Immunol. 126:1614-1619. Mailing List ------------ Please send comments (good and bad) regarding this program to: quality@valis.microbiol.washington.edu This will guarantee it gets to the appropriate person(s). We will also automatically put you on a list to be notified when a new version is available. Usage ----- Create an input data file using the included sample file as a guideline. 1) Each analysis starts with a line beginning with '#', followed by a description you assign to the analysis. The entire description line will be used, so you don't need to avoid spaces, or only use one word. 2) The next line contains the list of sample quantities at each dilution, each value separated by white space (spaces or tabs). 3) Line 3 is the corresponding list of PCR amplifications at each of the dilutions in step 2. 4) Line 4 is the corresponding list of positive PCR amplifications. If zero positives occurred, you MUST enter zero. This isn't a spreadsheet-- leaving a blank data point (such as two tabs) will slide these values to the left, resulting in an error message of "## Error: fewer PCR values than dilutions!" 5) probFalsePos and probFalseNeg can then be specified on separate lines. These lines have a bit more leeway, in that either Pos or Neg can appear first, and either or both can be omitted (and will default to 0.0). They MUST be spelled as shown, though, or they will not be recognized. It's probably easier just to leave them in, though. 6) Repeat steps 1-5 as necessary for the total analyses you wish to run. Blank lines will cause unexpected results, so don't include any. Strange results will also occur if you try reading text files that don't use the platform-appropriate endline characters (e.g. a Macintosh-formatted file into the PC version). Limitations ----------- - Max 50 dilutions per analysis (analyses per file are unlimited, though) Rather than code logic to handle any number of dilutions, this limit was chosen arbitrarily as higher than any reasonable experiment would require. Revision history ---------------- (M = Mac, P = PC, U = Unix, A = All versions) ** v1.1.6: - Added feature to calculate the volume of DNA per reaction to obtain user defined percent of positive PCR reactions. ** v1.1.5: - Added feature to calculate probability of not resampling. ** v1.1.4: A - Cleaned up and reorganized CFile objects to enable better maintenance of file handling code in the future. A - Binaries built using CodeWarrior Pro 4.1. ** v1.1.3 (not publicly released): A - Fixed a missed member initialization, which could cause crashes. Most noticeable in the Mac version, where it would crash at startup. M - Compiled w/ the newest version of CodeWarrior, w/ improved optimizations and file access. Therefore significantly faster. ** v1.1.2: A - Eliminated memory leaks A - Output for each analysis is now built up in memory, then sent to output. This is faster than sending output a line at a time, but causes a somewhat jerky impression. M - PowerPC optimizations were cranked as high as I could. (For some reason, this hadn't been done in previous PPC versions.) M - Output file now has a creator of 'R*ch' (BBEdit) rather than 'CWIE' (CodeWarrior). M - QUALITY's creator is now 'CBU1' so that unknown files on your drive whose creator is '????' do not claim to be "QUALITY documents". Such existing files will probably still make that claim until you rebuild your desktop. ** v1.1.1: A - First public release. If, for some reason, you have on older version, throw it away. M - The Macintosh version is no longer limited to 32k of output text. The Metrowerks SIOUX package has been modified to get around the limit of 32k of text inherent in the Macintosh's built-in text handling routines. (Unfortunately, this also increases the application size noticeably.) The default 1MB RAM allocation should accomodate almost any real-world data file. If the program stops execution prematurely on extraordinarily large data files, you should quit the program, assign it more memory in the Finder's "Get Info" window, then run the program again. P - The current .exe file was created using Metrowerks Codewarrior. It only runs under Win95 or WinNT. Open an MS-DOS window, then run it--if you just "Run" from the Start menu, etc., the window will close immediately after the program is finished. Running it from a DOS session will not do that. If anyone wishes to compile the source to run in raw DOS for us to distribute, please let me know. DISCLAIMER --------- This software is experimental and provided "as is" without obligation by the University of Washington to provide support of any kind. All risk as to the operation of the software and documentation and the possession and use of the software shall be with the recipient. 07jan98 Paul Goracke