Display Settings:

Items per page
We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

Results: 7

1.
Figure 5

Figure 5. From: Model-based analysis of two-color arrays (MA2C).

Workflow chart of MA2C. MA2C is fully automated and performs the tasks as shown.

Jun S Song, et al. Genome Biol. 2007;8(8):R178-R178.
2.
Figure 6

Figure 6. From: Model-based analysis of two-color arrays (MA2C).

Geometrical interpretation of the normalization method. Our method first subtracts the baseline from log intensity vectors within each GC bin and then projects the adjusted vectors onto v-axis, yielding log mean-scaled ratios of the Cy5 and Cy3 signals within each GC bin. Finally, the projected values are adjusted for variance.

Jun S Song, et al. Genome Biol. 2007;8(8):R178-R178.
3.
Figure 4

Figure 4. From: Model-based analysis of two-color arrays (MA2C).

Log-ratio density plots. All samples are from [12]: (a) raw data; (b) median adjusted data; (c) QQ normalized data; (d) Lowess normalized data; (e) MA2C (Simple) normalized data; (f) MA2C (Robust C = 2) normalized data. Different colors correspond to different samples.

Jun S Song, et al. Genome Biol. 2007;8(8):R178-R178.
4.
Figure 7

Figure 7. From: Model-based analysis of two-color arrays (MA2C).

Average intensities of the control channel data from [12] as a function of position-specific GC counts. Each 50-mer probe is partitioned into 5 equal parts of 10 nucleotides, and average intensities are computed as a function of GC counts in each part. Different colors represent different samples. The GC-related variations of intensities behave similarly across the five locations on probes, and we thus see that the GC effect is not position specific.

Jun S Song, et al. Genome Biol. 2007;8(8):R178-R178.
5.
Figure 1

Figure 1. From: Model-based analysis of two-color arrays (MA2C).

ChIP-chip. Regions of interest on DNA are densely tiled, with probes separated by short distances. In this figure, each bar corresponds to the log-ratio hybridization signals of two channels measured by a probe. Small sub-regions that are over-represented compared to the genomic background will appear as pronounced peaks (in this example, the middle peak represents the DNA fragments containing a protein-binding site). The computational challenge is to normalize the data properly and to detect confident enriched regions by filtering out false peaks (left and right peaks in this example).

Jun S Song, et al. Genome Biol. 2007;8(8):R178-R178.
6.
Figure 3

Figure 3. From: Model-based analysis of two-color arrays (MA2C).

Histograms of intensities. (a) Histogram of single-channel log-intensity values for a single array from 28256_Input [12]. The red bars represent the log-intensities for the probes with G+C less than 20, indicating that the bimodal behavior is caused by the GC content of probes. (b) Density plot of single channel log-intensities for two channels on the same array (28256_ChIP, black; 28256_Input, red). Notice that both the scale and the mean of the individual channels must be adjusted to properly normalize the arrays. (c) The raw data log-ratio values (28256_ChIP/28256_Input) for the same array in (b). Note that the 'bump' at 0 is not caused by enrichment but by lack of channel specific normalization of the data.

Jun S Song, et al. Genome Biol. 2007;8(8):R178-R178.
7.
Figure 2

Figure 2. From: Model-based analysis of two-color arrays (MA2C).

Scatter plots of the Cy5 versus Cy3 channels for 50-mer probes from [12] with (a) 28256_Input versus 28256_ChIP for G+C = 11 bases and (b) 28256_Input versus 28256_ChIP for G+C = 39 bases. The correlation is 0.364 in (a) and 0.860 in (b). (c) Plot of the inter-channel correlation (28256_Input, 28256_ChIP) across GC bins within the same array. The higher GC-count probes are more correlated and, therefore, should be more reliable in detecting differentially expressed or enriched probes. That is, in ChIP-chip, more than 99% of probes just measure the background and, thus, should ideally give similar results for the two channels. The correlation between the two channels, however, depends on the GC content of the probes. Since the two-channel correlation for high-GC probes is much higher than that for low-GC probes, significant two-channel fold-changes in the former category are much more reliable than those in the latter category, where large fold-changes may readily occur by chance.

Jun S Song, et al. Genome Biol. 2007;8(8):R178-R178.

Display Settings:

Items per page

Supplemental Content

Recent activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...
Write to the Help Desk