How is 'ESTIMATED_LIBRARY_SIZE' computed in picard tools?

Question

[abhishekniroula7]

Hi,

I used picard tools to mark duplicates in my paired RNAseq reads. Now, I am trying to understand the metrics obtained from the tool.

## METRICS CLASS    picard.sam.DuplicationMetrics
LIBRARY UNPAIRED_READS_EXAMINED READ_PAIRS_EXAMINED SECONDARY_OR_SUPPLEMENTARY_RDS  UNMAPPED_READS  UNPAIRED_READ_DUPLICATES    READ_PAIR_DUPLICATES    READ_PAIR_OPTICAL_DUPLICATES    PERCENT_DUPLICATION ESTIMATED_LIBRARY_SIZE
Unknown Library 6874536 70195365    21943811    13162992    6463873 28645311    30223   0.432923    60563946

Even after reading the documentation about MarkDuplicates and EstimateLibraryComplexity, I cannot recalculate the laste column (ESTIMATED_LIBRARY_SIZE). Is there a formula to calculate this value? Any explanation or suggestion where to find this information would be greatly appreciated.

Answer 1

[Pierre Lindenbaum]

From the picard code: https://github.com/broadinstitute/picard/blob/master/src/main/java/picard/sam/DuplicationMetrics.java#L115

    /**
     * Estimates the size of a library based on the number of paired end molecules observed
     * and the number of unique pairs observed.
     *
     * Based on the Lander-Waterman equation that states:
     *     C/X = 1 - exp( -N/X )
     * where
     *     X = number of distinct molecules in library
     *     N = number of read pairs
     *     C = number of distinct fragments observed in read pairs
     */
    public static Long estimateLibrarySize(final long readPairs, final long uniqueReadPairs) {
        final long readPairDuplicates = readPairs - uniqueReadPairs;

        if (readPairs > 0 && readPairDuplicates > 0) {
long n = readPairs;
(...)

[abhishekniroula7]

Thank you for the quick response.