Web Tables 1-12. Example of the UPGMA method of tree construction.

Web Table 7.1. Distance matrix of 13 globins using the Poisson correction. Here we prepare to cluster the two closest OTUs (10 and 11). The 13 globin sequences are given in the legend to Fig. 7.1, and the distance matrix of Fig. 7.xa is shown in this table. The least dissimilar pair that we will cluster first is indicated in red (distance = 0.07), and the distances of OTUs 10 and 11 to all the other OTUs are highlighted in yellow.

taxon	1	2	3	4	5	6	7	8	9	10	11	12	13
1
2	0.18
3	0.15	0.11
4	1.36	1.36	1.36
5	1.40	1.47	1.36	0.24
6	1.51	1.51	1.43	0.22	0.27
7	1.23	1.20	1.17	0.88	0.94	0.90
8	1.23	1.26	1.17	0.84	0.90	0.86	0.15
9	1.33	1.29	1.23	0.92	0.94	0.88	0.25	0.28
10	1.51	1.68	1.51	1.14	1.14	1.12	1.33	1.33	1.26
11	1.55	1.64	1.55	1.12	1.14	1.12	1.33	1.40	1.26	0.07
12	2.02	1.95	1.95	1.73	1.73	1.73	1.47	1.59	1.59	1.73	1.78
13	1.47	1.51	1.43	1.68	1.73	1.95	1.68	1.59	1.78	1.51	1.55	1.64

Web Table 7.2. First clustered matrix: OTUs 10 and 11 are combined, and we prepare to join OTUs 2 and 3. The two closest OTUs (10 and 11) have been clustered to create 10·11. The values in the newly joined 10·11 taxon are obtained for dissimilarity U_ab,c by averaging U_a,c and U_b,c. For example, for U_10·11,1 we take the average of U_10,1 and U_11,1. This means we inspect web table 7.1 and take the average of the distance of OTU 10 to OTU 1 (that is 1.51, highlighted in yellow) and OTU 11 to OTU 1 (that is 1.55, also highlighted in yellow), obtain the average value (1.53) and enter that in web table 7.2 (see value 1.53 in blue). When this process is complete, we identify the next smallest value; it is 0.11 (indicated in red) between OTUs 2 and 3. (Column 13 is blank and hence is deleted.)

taxon	1	2	3	4	5	6	7	8	9	10·11	12
1
2	0.18
3	0.15	0.11
4	1.36	1.36	1.36
5	1.40	1.47	1.36	0.24
6	1.51	1.51	1.43	0.22	0.27
7	1.23	1.20	1.17	0.88	0.94	0.90
8	1.23	1.26	1.17	0.84	0.90	0.86	0.15
9	1.33	1.29	1.23	0.92	0.94	0.88	0.25	0.28
10·11	1.53	1.66	1.53	1.13	1.14	1.12	1.33	1.365	1.26
12	2.02	1.95	1.95	1.73	1.73	1.73	1.47	1.59	1.59	1.755
13	1.47	1.51	1.43	1.68	1.73	1.95	1.68	1.59	1.78	1.53	1.64

Web table 7.3. Second clustered matrix: combining OTUs 2 and 3, and preparing to next combine OTUs 7 and 8. After calculating the means of 2·3 to each of the other values, we next identify the smallest distance (0.15 for OTUs 7 and 8), highlight that cell in red to mark its position, and highlight the rows and columns corresponding to OTUs 7 and 8 for which we will calculate the average values to be placed in Web table 7.4.

taxon	1	2·3	4	5	6	7	8	9	10·11	12
1
2·3	0.165
4	1.36	1.36
5	1.40	1.415	0.24
6	1.51	1.47	0.22	0.27
7	1.23	1.185	0.88	0.94	0.90
8	1.23	1.215	0.84	0.90	0.86	0.15
9	1.33	1.26	0.92	0.94	0.88	0.25	0.28
10·11	1.53	1.595	1.13	1.14	1.12	1.33	1.365	1.26
12	2.02	1.95	1.73	1.73	1.73	1.47	1.59	1.59	1.755
13	1.47	1.47	1.68	1.73	1.95	1.68	1.59	1.78	1.53	1.64

Web table 7.4. Third clustered matrix: OTUs 7,8 are combined, and we prepare to combine OTU 1 with OTU 2·3. After calculating the means of 2·3 to each of the other values, we next identify the smallest distance (0.165 for OTUs 1 and 2·3), highlight that cell in red to mark its position, and highlight the rows and columns corresponding to OTUs 1 and 2·3 for which we will calculate the average values to be placed in Web table 7.5.

taxon	1	2·3	4	5	6	7·8	9	10·11	12
1
2·3	0.165
4	1.36	1.36
5	1.40	1.415	0.24
6	1.51	1.47	0.22	0.27
7·8	1.23	1.20	0.86	0.92	0.88
9	1.33	1.26	0.92	0.94	0.88	0.265
10·11	1.53	1.595	1.13	1.14	1.12	1.348	1.26
12	2.02	1.95	1.73	1.73	1.73	1.53	1.59	1.755
13	1.47	1.47	1.68	1.73	1.95	1.635	1.78	1.53	1.64

Web table 7.5. Fourth clustered matrix: OTU 1 is combined with OTU 2·3, and we prepare to combine OTUs 4 and 6.

taxon	(1, 2·3)	4	5	6	7·8	9	10·11	12
(1,2·3)
4	1.36
5	1.408	0.24
6	1.49	0.22	0.27
7·8	1.215	0.86	0.92	0.88
9	1.295	0.92	0.94	0.88	0.265
10·11	1.563	1.13	1.14	1.12	1.348	1.26
12	2.02	1.73	1.73	1.73	1.53	1.59	1.755
13	1.985	1.68	1.73	1.95	1.635	1.78	1.53	1.64

Web table 7.6. Fifth clustered matrix: OTUs 4 and 6 are combined, and we prepare to combine the next closest pair, that is OTU 4·6 with OTU 5.

taxon	(1, 2·3)	4·6	5	7·8	9	10·11	12
(1,2·3)
4·6	1.425
5	1.408	0.255
7·8	1.215	0.87	0.92
9	1.295	0.90	0.94	0.265
10·11	1.563	1.125	1.14	1.348	1.26
12	2.02	1.73	1.73	1.53	1.59	1.755
13	1.985	1.815	1.73	1.635	1.78	1.53	1.64

Web table 7.7. Sixth clustered matrix: OTUs 4·6 and 5 are combined, and we prepare to combine OTU 7·8 with OTU 9.

taxon	(1, 2·3)	(4·6,5)	7·8	9	10·11	12
(1,2·3)
(4·6,5)	1.4165
7·8	1.215	0.895
9	1.295	0.92	0.265
10·11	1.563	1.133	1.348	1.26
12	2.02	1.73	1.53	1.59	1.755
13	1.985	1.773	1.635	1.78	1.53	1.64

Web table 7.8. Seventh clustered matrix: OTUs 7·8 and 9 are combined, and we prepare to combine OTU (7·8,9) with OTU (4·6,5).

taxon	(1, 2·3)	(4·6,5)	(7·8,9)	10·11	12
(1,2·3)
(4·6,5)	1.4165
(7·8,9)	1.255	0.908
10·11	1.563	1.133	1.304
12	2.02	1.73	1.56	1.755
13	1.985	1.773	1.708	1.53	1.64

Web table 7.9. Eighth clustered matrix: OTUs (7·8,9) and (4·6,5) are combined, and we prepare to combine OTU (4·6,5) (7·8,9) with OTU 10·11.

taxon	(1, 2·3)	(4·6,5) (7·8,9)	10·11	12
(1,2·3)
(4·6,5) (7·8,9)	1.336
10·11	1.563	1.219
12	2.02	1.645	1.755
13	1.985	1.741	1.53	1.64

Web table 7.10. Ninth clustered matrix: OTUs (4·6,5) (7·8,9) and 10·11 are combined, and we prepare to combine OTU (1, 2·3) with OTU [(4·6,5) (7·8,9)](10·11).

taxon	(1, 2·3)	[(4·6,5) (7·8,9)](10·11)	12
(1,2·3)
[(4·6,5) (7·8,9)](10·11)	1.449
12	2.02	1.700
13	1.985	1.636	1.64

Web table 7.11. Tenth clustered matrix: OTUs (1, 2·3) and [(4·6,5) (7·8,9)](10·11) are combined, and we prepare to combine OTU 12 with OTU 13.

taxon	(1,2·3)[[(4·6,5) (7·8,9)](10·11)]	12
(1,2·3)[[(4·6,5) (7·8,9)](10·11)]
12	1.86
13	1.811	1.64

Web table 7.12. Eleventh clustered matrix: OTUs 12 and 13 are combined, and we prepare to combine OTU x with OTU x. To obtain the value 1.836 we take the average of 1.86 and 1.811 from web table 7.11.

taxon	(1,2·3)[[(4·6,5) (7·8,9)](10·11)]	12·13
(1,2·3)[[(4·6,5) (7·8,9)](10·11)]
12·13	1.836