PSMB_Seqan_2013_NGS_Quality_Control_Details_Features

Our application will read fastq files and collect statistical data about its content. The input itself will not be modified. A summary document will be created to display the results.

The application implements 3 areas of functionality: input, data collection, results output formatting. Each functionality is classified as Must-Have or Nice-To-Have. Implementation and testing will be done by Antje A or Daniel D

Input

The application will be able to

1. read fastq files Must-Have A
2. read bam files Nice-To-Have D
3. read compressed fastq files Nice-To-Have D

Data Collection

The following data will be collected:

1. input filename and format Must-Have A
2. which scoring system was used Must-Have A
3. conversion of scoring systems Must-Have D
4. total number of sequences Must-Have A
5. overall quality score average of all bases in all sequences Must-Have A
6. overall GC percent of all bases in all sequences Must-Have A
7. per read and per position: Must-Have A
   • basic quality distribution data: median, mean, quantiles (10,25,75,90)
   • distribution of [A,C,G,T]
   • GC percent content
   • N Content
8. for all reads: Must-Have A
   • mean qualities distribution A
   • sequence length distribution A
9. overall sequence metrics
   • duplicated sequences Nice-To-Have A
   • k-mer distribution Must-Have D

Summary Document Generation

• Primary application output will be a tab-separated text file. A
• The tsv file can be read by an accompanying R-script. This R-script and an HTML document will show the graphics output generated by R from the data. D
• Document generation itself will be performed by a secondary application. D

Application Options

• force quality score system
• k-mer length
• quick analysis (randomly choose a subset of reads to analyze)

Bonus Bonus Bonus List

• linking with Galaxy or KNIME
• One Script wich starts all the other
• Graphical User Interface

Milestones

1st week

testing and implementation of a functionally minimal version that works through all steps

2nd week

testing an implementation of all basic statistics (A) and k-mer content (D)

3rd week

testing and implementation of sequence duplication (A) and output renement (D)

4th week

buffer for surprises, testing and implementation of NICE-TO-HAVE features (A+D)