Team:Nevada/Interpolated Data
From 2010.igem.org
(→Boolean Input Functions) |
(→Boolean Input Functions) |
||
(One intermediate revision not shown) | |||
Line 551: | Line 551: | ||
== Boolean Input Functions == | == Boolean Input Functions == | ||
- | When reading the input functions, the first number in a line gives the number of inputs for that gene. Subsequent numbers, in parentheses, indicate the genes contributing to the true function. Logical operators are: ( - = NOT ), ( & = AND ), ( | = OR ). | + | These functions represent all possible interactions that could be mathematically modeled, given the input data. When reading the input functions, the first number in a line gives the number of inputs for that gene. Subsequent numbers, in parentheses, indicate the genes contributing to the true function. Logical operators are: ( - = NOT ), ( & = AND ), ( | = OR ). |
Gene numbers corresponding to the 8 genes are as follows: | Gene numbers corresponding to the 8 genes are as follows: | ||
Line 580: | Line 580: | ||
- | For | + | For this data set, inputs were limited to 4 per gene. This, in conjunction with increasing the number of time points to take values from, dramatically reduced the number of Boolean input functions in all cases. However, ICE1 and RAP2.6 were still associated with millions of functions. Both of these genes share identical expression profiles ( "off" at all time points ) and identical input functions. For ICE1, this may be due to post-transcriptional regulation and being active even though transcript levels remain constant under cold-stress, but it is unknown if RAP2.6 is post-transcriptionally modified. Because of the unwieldy number of input functions for ICE1 and RAP2.6, data has again been omitted for these genes. |
- | Note also that DREB1A, ZAT6 and RAP2.1 do not appear in the list, as no Boolean input functions exist that generate the data in our dataset. | + | Note also that DREB1A, ZAT6 and RAP2.1 do not appear in the list, as no Boolean input functions exist that generate the data in our dataset. All three genes are involved in a number of possible Boolean input functions for other genes. |
Latest revision as of 03:32, 28 October 2010
Contents |
Interpolation Method
Natural cubic splines were made based on the four time points for each gene. This produced 3 polynomials, for each gene, defining their time states: one for the interval from 0 to 3 hours, the second from 3 to 6 hours and the third from 6 to 24 hours. For all genes, the third polynomial was also used to extrapolate out to 30 hours, in order to get a sense of whether or not some of these genes may have a delayed response to freezing. 16 time points were taken from the 30-hour interpolated/extrapolated data at 2 hour intervals to produce this data set, i.e. ( t0 = 0 hours, t1 = 2 hours, t2 = 4 hours...t15 = 30 hours ).
Interpolated Data Set
Raw Data | ICE1 | DREB1A | DREB1B | DREB1C | rd29A | ZAT6 | RAP2.1 | RAP2.6 |
t0 | 1358.1 | 96.6 | 24.5 | 69.4 | 1509.2 | 145.3 | 100.4 | 58.5 |
t1 | 1450.6 | 4600.5 | 3748.3 | 9051.6 | 2125.8 | 455.4 | 78.4 | 58.2 |
t2 | 1515 | 6617.4 | 5072.8 | 12724.1 | 5330.4 | 965.3 | 166.6 | 55.8 |
t3 | 1547.9 | 6181.2 | 4099 | 11249.3 | 10644.4 | 1620.6 | 357.8 | 52.3 |
t4 | 1566.7 | 5400.5 | 2970.9 | 9273.8 | 15161.9 | 2168.8 | 549.6 | 51.6 |
t5 | 1576.1 | 4709.9 | 2112.2 | 7731.3 | 18402.6 | 2571.8 | 722.5 | 53.9 |
t6 | 1577.2 | 4098.1 | 1489.2 | 6567.6 | 20526.3 | 2847.7 | 878.6 | 59 |
t7 | 1571.2 | 3554 | 1068.2 | 5728.6 | 21692.4 | 3014.6 | 1020.5 | 66.4 |
t8 | 1559.4 | 3066.1 | 815.6 | 5160.2 | 22060.6 | 3090.8 | 1150.5 | 75.8 |
t9 | 1542.8 | 2623.4 | 697.7 | 4808.4 | 21790.5 | 3094.3 | 1271.1 | 86.8 |
t10 | 1522.6 | 2214.4 | 680.8 | 4618.8 | 21041.6 | 3043.4 | 1384.4 | 98.9 |
t11 | 1500.2 | 1828 | 731.3 | 4537.6 | 19973.6 | 2956.1 | 1493.7 | 111.8 |
t12 | 1476.5 | 1452.8 | 815.4 | 4510.4 | 18745.9 | 2850.7 | 1599.3 | 125 |
t13 | 1452.8 | 1077.6 | 899.5 | 4483.2 | 17518.2 | 2745.3 | 1705.5 | 138.3 |
t14 | 1430.4 | 691.2 | 950 | 4402 | 16450.2 | 2658 | 1814.2 | 151.1 |
t15 | 1410.2 | 282.2 | 933.1 | 4212.4 | 15701.3 | 2607.1 | 1927.5 | 163.2 |
Log Base 2 of Normalized Ratios
Values were normalized to 1 at t0, and log base 2 values were taken.
Log2 Ratios | ICE1 | DREB1A | DREB1B | DREB1C | rd29A | ZAT6 | RAP2.1 | RAP2.6 |
t0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
t1 | 0.0951 | 5.5736 | 7.2573 | 7.0271 | 0.4942 | 1.6481 | -0.3568 | -0.0074 |
t2 | 0.1577 | 6.0981 | 7.6939 | 7.5184 | 1.8205 | 2.7319 | 0.7306 | -0.0682 |
t3 | 0.1887 | 5.9997 | 7.3863 | 7.3407 | 2.8182 | 3.4794 | 1.8334 | -0.1616 |
t4 | 0.2061 | 5.8049 | 6.9220 | 7.0621 | 3.3286 | 3.8998 | 2.4526 | -0.1811 |
t5 | 0.2148 | 5.6075 | 6.4298 | 6.7996 | 3.6081 | 4.1457 | 2.8472 | -0.1182 |
t6 | 0.2158 | 5.4068 | 5.9256 | 6.5643 | 3.7656 | 4.2927 | 3.1294 | 0.0123 |
t7 | 0.2103 | 5.2013 | 5.4463 | 6.3671 | 3.8453 | 4.3749 | 3.3454 | 0.1827 |
t8 | 0.1994 | 4.9882 | 5.0570 | 6.2163 | 3.8696 | 4.4109 | 3.5184 | 0.3738 |
t9 | 0.1840 | 4.7633 | 4.8318 | 6.1145 | 3.8518 | 4.4125 | 3.6622 | 0.5693 |
t10 | 0.1649 | 4.5187 | 4.7964 | 6.0564 | 3.8014 | 4.3886 | 3.7854 | 0.7575 |
t11 | 0.1436 | 4.2421 | 4.8996 | 6.0308 | 3.7262 | 4.3466 | 3.8951 | 0.9344 |
t12 | 0.1206 | 3.9107 | 5.0567 | 6.0222 | 3.6347 | 4.2942 | 3.9936 | 1.0954 |
t13 | 0.0972 | 3.4797 | 5.1983 | 6.0134 | 3.5370 | 4.2399 | 4.0864 | 1.2413 |
t14 | 0.0748 | 2.8390 | 5.2771 | 5.9871 | 3.4462 | 4.1932 | 4.1755 | 1.3690 |
t15 | 0.0543 | 1.5466 | 5.2512 | 5.9236 | 3.3790 | 4.1653 | 4.2629 | 1.4801 |
Boolean Values (Threshold of 3.5)
Values from the log base 2 table which were greater than or equal to a 3.5-fold increase were assigned values of 1 ( "on" ) and values under 3.5 were assigned values of 0 ( "off" ).
Boolean Values | ICE1 | DREB1A | DREB1B | DREB1C | RD29A | ZAT6 | RAP2.1 | RAP2.6 |
t0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
t1 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
t2 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
t3 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
t4 | 0 | 1 | 1 | 1 | 0 | 1 | 0 | 0 |
t5 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 |
t6 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 |
t7 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 |
t8 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 |
t9 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 |
t10 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 |
t11 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 |
t12 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 |
t13 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 0 |
t14 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 0 |
t15 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 0 |
Boolean Input Functions
These functions represent all possible interactions that could be mathematically modeled, given the input data. When reading the input functions, the first number in a line gives the number of inputs for that gene. Subsequent numbers, in parentheses, indicate the genes contributing to the true function. Logical operators are: ( - = NOT ), ( & = AND ), ( | = OR ).
Gene numbers corresponding to the 8 genes are as follows:
1 = ICE1
2 = DREB1A
3 = DREB1B
4 = DREB1C
5 = rd29A
6 = ZAT6
7 = RAP2.1
8 = RAP2.6
As an example of a true Boolean function, consider:
4 ( -1 | 3 ) & ( 5 ) & ( 6 )
This means that the selected gene has four inputs giving a true value, and those inputs are ( ( NOT ICE1 ) OR DREB1B ) AND ( rd29A ) AND ( ZAT6 ).
For this data set, inputs were limited to 4 per gene. This, in conjunction with increasing the number of time points to take values from, dramatically reduced the number of Boolean input functions in all cases. However, ICE1 and RAP2.6 were still associated with millions of functions. Both of these genes share identical expression profiles ( "off" at all time points ) and identical input functions. For ICE1, this may be due to post-transcriptional regulation and being active even though transcript levels remain constant under cold-stress, but it is unknown if RAP2.6 is post-transcriptionally modified. Because of the unwieldy number of input functions for ICE1 and RAP2.6, data has again been omitted for these genes.