DNA sequence (H) curves of the human immunodeficiency virus 1 and some related viral genomes

The complete nucleotide sequences of several human immunodeficiency virus 1 (HIV-1) genomes were converted by computer to respective H curves. These three-dimensional space curves embody all the information contained in the sequence due to their abstract vectorial structure. For one sequence (HIV-1 isolate BRU) special efforts were made to maximize the available resolution (the number of nucleotides visually discernible within a unit length of the curve) when making a hard, master copy of the H curve. Using a computergraphic/photographic hybrid process the 9191 nucleotides of this HIV-1 sequence were condensed into an H curve of only 37.1 cm vertical length. Although each 1-mm segment of this curve represented 25 nucleotide residues, each of the individual nucleotides of the entire sequence was still distinguishable upon direct inspection using a simple magnifying lens. Several functionally important loci of the HIV-1 sequence could be recognized on the H curve owing to characteristic line forms at corresponding locations. Utilizing H curves of lower resolution, the total nucleotide sequences of several different HIV-1 isolates and related viral sequences [Visna, equine infectious anemia (EIAV), Moloney murine leukemia (Mo-MLV), bovine leukemia (BLV), and human T-cell leukemia, type I (HTLV-I)] were visually compared side by side. An interesting similarity was noted between the location of the S3 fragment of the EIAV sequence and that of a relatively G-C rich region on the env portion of the HIV-1 sequences.