Table - link
||Krogh A, Larsson B, von Heijne G, Sonnhammer EL. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol. 2001 Jan 19 305(3):567-80. p.573 table 4PubMed ID11152613
||P.568 left column 3rd paragraph:"Here [investigators] describe a new method, TMHMM, based on a hidden Markov model (HMM) approach (a preliminary description of TMHMM has been published by Sonnhammer et al., 1998). It resembles the method described by Jones et al. (1994), in that it has specialized modeling of various regions of a membrane protein: helix caps, middle of helix, regions close to the membrane, and globular domains."
||P.572 right column bottom paragraph:"Estimates of the percentages of all annotated genes that encode integral membrane proteins of the helix bundle class are presented in Table 4. In general, these vary between ∼20% and ∼30%. A previously suggested correlation between the frequency of predicted membrane proteins and the total number of genes in the genome (Wallin & von Heijne, 1998) is not discernible in these new estimates (see Figure 6). Two organisms have noticeably higher fractions of membrane proteins than the general trend: Plasmodium falciparum and C. elegans." See note beneath table