Daniel:Notebook/GenomeMiner

From ZhangLabWiki
Revision as of 22:41, 30 August 2013 by >Djacobse (→‎HL155)
Jump to navigation Jump to search

Back to Main

Link to useful Perl website: Basic Perl

Calendar

<calendar> name=Daniel:Notebook/GenomeMiner format=%name/%year-%month-%day date=2013/02/01 view=oneyear </calendar>

Useful Linux Commands

SCP to laptop

%% scp <remote user>@<remote machine>:<remote path> <local file>
scp djacobse@132.239.135.41:/home/djacobse/HL155/130628_HL155/probecount.csv probecount.csv

HL155

Analysis of 3 probe production companies: MYcroarray, Agilent, and Custom Array.

Data Analysis Spreadsheet: File:ProbeComparison.xlsx

MATLAB Master File: HL155_Master.txt

Raw Data

Raw Data Methods: (Dan,Matt)

Probe Set Total reads Filtered reads Perfect match 1 ins/del 1 sub 2 ins/del 2 sub 1 ins/del & 1 sub 3+
Mycroarray 20727275 15082383 4411040 3584377 2417322 1893421 754110 1647242 374871
Agilent 0 gap 4953642 4448339 3294962 102309 808922 12240 129671 24455 75780
Agilent 20 gap 4052727 3578431 2676974 101104 619290 14445 92296 20184 54138
Agilent 9006369 8026770 5971936 203413 1428212 26685 221967 44639 129918
Custom Array 12K 5195976 4533981 3176472 77874 962022 4934 178599 30019 104061

As Percent

Probe Set Total reads Filtered reads (%) Perfect match (%Filtered) 1 ins/del (%Filtered) 1 sub (%Filtered) 2 ins/del (%Filtered) 2 sub (%Filtered) 1 ins/del & 1 sub (%Filtered) 3+ (%Filtered)
Mycroarray 20727275 72.8 29.2 23.8 16.0 12.6 5.0 10.9 2.5
Agilent 0 gap 4953642 89.8 74.1 2.3 18.2 0.3 2.9 0.5 1.7
Agilent 20 gap 4052727 88.3 74.8 2.8 17.3 0.4 2.6 0.6 1.5
Agilent 9006369 89.1 74.4 2.5 17.8 0.3 2.8 0.6 1.6
Custom Array 12K 5195976 87.3 70.1 1.7 21.2 0.1 3.9 0.7 2.3

Comparison

Comparison of Errors

File:Probecompare HL155.png


Comparing errors per 100 bp

  Insertions/Deletions (/100 bp) Substitutions (/100 bp)
Mycroarray 1.20 0.74
Agilent 0.08 0.48
Custom Array 12K 0.05 0.60

Comparison by Error

Y axis is given in % chance, so Agilent probes have overall error rate of ~2%, or 1 in 50 bp.

File:ProbeComparison Improved.png

Probe Counts Comparison

Looking at the comparison between counts of the three companies. Bar graphs below divide the probes into 4 categories: zero, under-amplified, normally amplified, and over-amplified. All counts are normalized to the expected number of reads. Expected number of reads is calculated by dividing the total number of reads by the total number of probes in the set. The expected number of reads is therefore the number of reads if all probes were amplified equally. Under amplified probes are defined as those with less than half the number of expected reads. This is because PCR amplifies at 2^n. Having half as many reads is therefore the same as the probe being amplified one less cycle than the norm. Similarly, over-amplified is defined as having more than twice the number of reads. Again, having twice as many reads is the same as saying the probe was amplified for one extra cycle.

The four categories are:

Zero: The probe was never counted

Under Amplified: probe count < 0.5*expected

Normally Amplified: 0.5*expected ≤ probe count ≤ 2*expected

Over-amplified: probecount > 2*expected

Probe Count Comparison

File:Probecompare Bargraph HL155 08302013.png