A) DNA sequencing

1) Manually

2) Automated

Dye primer chemistry

Dye terminator chemistry

3) Problems in the sequencing reaction(s)

B) Computer analysis of sequenced data

A) DNA Sequencing: the dideoxy chain-termination method of Sanger is still the preferred method.

1) Manual sequencing- laborious, two-day minimum to complete, max of 4-6 different samples sequenced per two days, the terminating nucleotide is labeled with a radioactive tracer.

2) Automated sequencing- easy, reactions are completed from 12-24 hours, can run hundreds of reactions per day, the primer (dye primer chemistry) or terminating nucleotides (dye terminator chemistry) are labeled with fluorescent dyes.

-In the dye terminator chemistry each of the four terminating nucleotide is labeled with a different dye. Therefore, the entire sequencing reaction can be run as one. A laser excites the fluorescent dyes as they come out of the gel and detectors collect the emission intensities of the four different wavelengths and generates a chromatogram.

3) Problems in the sequencing reaction(s): imperfections in the sequencing reaction, gel electrophoresis.

For example: the 1^st 50 or so peaks of a trace are noisy and unevenly spaced due to greater effects of the dye and specificity of base sequences on mobility. At the end of a reaction, diffusion effects increase, mass difference of the base sequence decrease, and the number of labeled fragments decreases. Also, in the dye terminator chemistry method, the polymerase has a reduced affinity for the dye labeled nucleotide, especially for a labeled G following and A.

B) Computer analysis of sequences: converts the gel image to an inferred sequence.

Four phases: 1- lane tracking

Use for identifying lane boundaries

2- lane profiling

Each of the four signals is summed across the lane width to create a profile

3-trace processing

Smoothes and deconvolve the trace, reduce noise and account for errors in the sequencing reactions.

4- base-calling

Translates the processed trace into a sequence of bases (Phred)

A base-calling program for automated sequencer traces. It assigns an error probability to each base called and yields 40-50% lower errors than the ABI software: Input are in the form of chromatogram files in ABI format or standard chromatogram format (SCF).

Base-calling pipeline: consists of four-phases

1. It determines the ideal or predicted peak locations using the fact that fragments should be locally evenly spaced. It then uses this information to determine the correct number of bases in a region where peaks are not well resolved, noisy or displaced.
2. Observed peaks are identified in the trace
3. Observed peaks are matched to predicted peak location, omitting some and splitting other. This is the main base sequence.
4. Unmatched peaks are analyzed to see if they represent bases. If they represent bases, the base is inserted into the sequence.

Quality scores (Q) are based on probability errors (P).

Q=-10 log10 (P)

The phragmet assembly program is used to assemble shotgun DNA sequences accurately using Phred’s quality score files.

The following is just a very brief and incomplete description of how PHRAP generates assemblies. For more detailed information, please check the "Additional Readings" section.

1. PHRAP reads the sequences from the input file, and quality values from the corresponding quality file (if present).

2. PHRAP masks likely "garbage" sequence at the beginning and end of reads, which is converted to 'N's. Specifically, PHRAP looks for regions that consist almost entirely of a single base; such regions are most likely due to poor data quality, and could give rise to spurious matches.

3. PHRAP identifies all potentially overlapping pairs of sequences. Two sequences must have at least one "word" of length minmatch (typically 14 bases) in common, and the alignment between the sequences must have an alignment score of at least minscore (default: 30).

4. PHRAP calculates modified quality scores for each base in each read, taking into account confirmation by overlapping reads as well as orientation and sequencing chemistry. PHRAP also calculates "LLR" scores for each pairwise alignment. LLR scores are a measure of overlap length and quality; high quality discrepancies that might indicate different copies of a repeat lead to low LLR scores. Potential problem clones like chimeras and deletions clones are also identified.

5. Merge reads into contigs, starting at the pairwise overlaps with the highest LLR scores. Probable deletion clones and chimeras are not used in any merges.

6. PHRAP calculate the contig ("consensus") sequences, which are pieced together from individual sequence reads with the highest adjusted quality scores at any base. Since the adjusted qualities take confirmation and discrepancies in other reads into account, the quality of the consensus sequence set to be identical to the revised quality of the highest quality read. If the original quality values were linked to error probabilities (like PHRED quality values are), the quality scores of the contig sequence are conservative estimates of the error probabilities in the contig sequences.

PHRAP typically reads input sequences from a single file, which must be in FASTA format. PHRAP will also read base-specific quality values, if it can find a file that has ".qual" appended to the name of the sequence file (for example quality files generated by PHRED or PHD2FASTA).

Any non-alphabetic characters, including '*', in the sequence lines are automatically stripped out when the file is read in, except for '-' which is converted to 'N'. The character '>' must not appear anywhere within the sequence, since it is assumed to start the header of a new sequence, even if it is not at the beginning of a line. Lower case letters are converted to upper case on readin, so it is no longer possible (as in previous PHRAP versions) to use case as a crude indication of quality.

PHRAP uses information about sequencing chemistry (dye primer or dye terminator) when calculating consensus quality values: confirmation of a dye primer read with a dye terminator counts as much as confirmation by an opposite-strand read.

If sequence from both ends of a clone is present in the assembly, this information is currently only used for consistency checking after assembly, not during the assembly itself (some independently developed finishing tools can also utilize the double-ended information).

After generating a FASTA file with all the individual sequence reads that are to be used in the assembly (for example by running PHRED and then CROSS_MATCH for vector screening), you are ready to run PHRAP. Let us assume that you the sequence file is called Reads.fasta, and that it is located in a directory called C:\Project\Asm1. Open an MS-DOS window, and type the following commands:

Depending on the project size, PHRAP will run for a few seconds to several hours (see next section). PHRAP will generate a number output files:

readslog.txt - a file with a lot of information about the assembly.

The following table give an overview of the files PHRAP creates:

The amount of memory (RAM) PHRAP needs depends on both the project size and the number of repeats in a project. For cosmid-sized projects with several hundred sequences, PHRAP may need 25 MB of RAM, and complete the assembly in less than three minutes on a 300 MHz Pentium II computer (cosmids with many repeats may need more memeory and time). On Unix workstations, PHRAP has successfully been used to assemble more than 30,000 reads; such large runs may require 4 GB of memory, and several days to complete.

The actual amount of RAM available on the computer that PHRAP is running on should be larger than the amount of memory PHRAP needs; for example, computer for cosmid assembly should have at least 32 MB of RAM. If the amount of memory PHRAP needs is not available as RAM, PHRAP's performance can be drastically slower, since the computer will spend most of the time swapping block of memory to and from the hard disks.

For optimal performance on larger projects, we suggest that you quit all open applications before starting PHRAP.

In the Windows and Unix versions, PHRAP uses command line options to control input, processing, and output. The command line options are delimited by a dash, "-".

The most commonly used command line options for PHRAP are: