Frame lengths and shifts

In addition to the length of window, we also have to consider how often we should calculate the spectrum. The frame shift is defined as the distance between two consecutive frames. Again, there is no single correct value for this. In some applications (e.g. 

In speech synthesis, it is common to perform pitch-synchronous analysis, in which the frame is centred around the pitch period. Since pitch is generally changing, this makes the frame shift variable. Pitch-synchronous analysis has the advantage that each frame represents the output of the same process, that is, the excitation of the vocal tract with a glottal pulse. In unvoiced sections, the frame rate is calculated at even intervals. Of course, for pitch-synchronous analysis to work, we must know where the pitch periods actually are this is not a trivial problem and wftl be addressed further in Section 12.7.2. Note that fixed-frame-shift analysis is sometimes referred to as pitch-asynchronous analysis. 

---

In our experience, the quality of the primers used for library amplification is the key to success. In many cases, the required upstream primers will be in excess of 60 bases in length, and we recommend that these should be ordered from providers offering premium quality synthesis and purification. Libraries generated using low quafity primers will include a higher proportion of truncation and frame shift mutants, wasting valuable library space, and potentially having an adverse effect on the outcome of selection experiments (see Section 2.2 below). 

These experiments make it clear that removing competition with release factors leads to more efficient incorporation of the desired amino acid. Unfortunately, the technology to incorporate nonstandard nucleotides into mRNAs coding for full-length proteins is not yet available. Alternatives that have been tested include using (i) a 4-nucleotide codon-anticodon pair, dubbed frame-shift suppression (Sect. 6.1), (ii) a rare codon, and (iii) cell-free extracts from organisms that are either deficient in a release factor (Sect. 5.1) or unable to translate one or more codons (Sect. 6.2). 

Peptide scanning is a very useful tool for screening a known protein sequence for active regions (e.g., epitopes). It was first described by Geysen and coworkers for peptides bound to polypropylene rods (28). The peptides are generated by shifting a frame of a distinct peptide length of a protein sequence of interest... 

---