Dear Modellers,

When running the compare.log script on the TvLDH example I obtained a qualitatively different aliignment than the "known" which I downloaded from the web-site. Here is an excerpt the compare.log file from the web-site , at line 442.

80 0.310 0.106 A 80 L 80 L 80 V 123 81 0.180 0.050 * D 81 D 81 D 81 D 124 82 0.212 0.039 Y 82 V 82 V 82 W 125 83 0.206 0.047 * A 83 A 83 A 83 A 126 84 0.165 0.073 L 84 I 84 I 84 L 127 85 0.122 0.037 * L 85 L 85 L 85 L 128 86 0.158 0.052 V 86 V 86 V 86 I 129 87 0.347 0.031 * G 87 G 87 G 87 G 130 88 0.609 0.174 A 88 S 88 S 88 A 131 # ============================================= 89 1.375 0.314 P 90 M 89 M 89 P 133 # ============================================= 90 1.479 0.649 R 97 L 100 D 99 R 140 # ============================================= 91 1.463 0.278 V 103 A 103 A 103 I 146 92 0.782 0.226 * N 104 N 104 N 104 N 147 93 1.087 0.426 G 105 V 105 V 105 G 148 94 0.845 0.340 K 106 K 106 K 106 Q 149 95 0.411 0.068 * I 107 I 107 I 107 I 150 96 0.386 0.096 * F 108 F 108 F 108 F 151 97 0.542 0.245 T 109 K 109 K 109 A 152 98 0.478 0.214 E 110 C 110 C 110 D 153 99 0.216 0.029 * Q 111 Q 111 Q 111 Q 154

Here is the same region of the file from my own run:

80 0.312 0.109 A 80 L 80 L 80 V 123 81 0.185 0.053 * D 81 D 81 D 81 D 124 82 0.216 0.042 Y 82 V 82 V 82 W 125 83 0.209 0.049 * A 83 A 83 A 83 A 126 84 0.167 0.075 L 84 I 84 I 84 L 127 85 0.126 0.040 * L 85 L 85 L 85 L 128 86 0.163 0.054 V 86 V 86 V 86 I 129 87 0.349 0.027 * G 87 G 87 G 87 G 130 88 0.610 0.174 A 88 S 88 S 88 A 131 # ============================================= 89 1.383 0.320 P 90 M 89 M 89 P 133 # ============================================= 90 1.481 0.649 R 97 L 100 D 99 R 140 # ============================================= 91 1.160 0.416 L 101 K 102 K 102 L 144 # ============================================= 92 1.456 0.284 V 103 A 103 A 103 I 146 93 0.776 0.235 * N 104 N 104 N 104 N 147 94 1.083 0.431 G 105 V 105 V 105 G 148 95 0.840 0.344 K 106 K 106 K 106 Q 149 96 0.407 0.072 * I 107 I 107 I 107 I 150 97 0.380 0.101 * F 108 F 108 F 108 F 151 98 0.536 0.252 T 109 K 109 K 109 A 152

If I understand the output correctly the gap structure of the alignments is different between the two runs. I will be sending my directory contents under separate cover. I would greatly appreciate your input as to whether this discrepancy is due to

A. Numerical differences between running the program on different machines. B. A problem with the Version for the Alpha. C. A problem with my installation of the Alpha Version. D. An error on my part running the program.

Thanks and best wishes, Rich -------------------------------------------------------------- Richard A. Friedman, PhD Associate Research Scientist Herbert Irving Comprehensive Cancer Center Oncoinformatics Core Lecturer Department of Medical Informatics Box 95, Room 130BB or P&S 1-420C Columbia University 630 W. 168th St. New York, NY 10032 (212)305-6901 (5-6901) (voice) friedman@cancercenter.ccc.columbia.edu http://cancercenter.ccc.columbia.edu/~friedman/

In Memoriam, Robert Kanigher