Analysis Rules/Thresholds

FaSTR™ DNA works by applying a set of fully configurable rules to streamline the analysis of STR DNA profiles. Beyond initial peak calling, FaSTR™ DNA aims to distinguish true signals (e.g., allelic peaks) from artefactual signals, such as stutter, pull-up, and dye blobs. FaSTR™ DNA allows the analyst to assess the peaks that have been called in raw DNA profile data and determine whether they should be retained (likely genuine), classified as stutter, or removed (likely artefactual).

When spectral pull-up peaks are detected via the analysis rules in the new version of FaSTR™ DNA, a marker is now displayed on the EPG to indicate the color of the peak causing the pull-up. This allows the analyst to easily visualize where one dye color is consistently causing pull-up across a whole dye channel. Additionally, the new cross channel artefact detection rule within FaSTR™ DNA is designed to bring the analyst’s attention to peaks that appear in the same position across multiple channels at once which are likely artefacts of the capillary electrophoresis rather than legitimate peaks.

Sample Quality and Integrity

The uncompromising adherence to quality and standards is mandatory and routinely part of the forensic DNA analysis Standard Operating Procedure and workflows. The new version of FaSTR™ DNA assists this process by automating the detection of primer flare signal (at the lower scan points beyond the analysis range).

In negative control samples, the presence of primer flare in all dye channels is verified and automatically annotated to each sample as a comment. In all sample types (excluding allelic ladder), the absence of primer flare in any dye channel triggers a sample warning status and is automatically annotated to each sample as a comment. The sample warning status remains even if all other analysis warnings (peak flags, other analysis rule warnings, etc.) have been resolved.

All sample quality comments are captured in the analysis summary report (PDF) for the analysis project, or viewable in the Sample Details information for each sample.

Stutter Model and Filter

For high template DNA profiles from a single contributor, distinguishing between allelic peaks and stutter artefacts is relatively straightforward. This becomes more complex when interpreting mixed DNA profiles where allelic peaks from a minor contributor can be of similar height to stutter artefacts from a major contributor since they look morphologically identical.

To detect and mark stutter peaks more accurately, FaSTR™ DNA can use, among other models, an allele-specific stutter model and filter. In this method, the stutter filter adjusts dynamically according to the empirically modelled expected stutter ratio. FaSTR™ DNA’s ability to detect and filter stutter enables more efficient initial manual interpretation of the profile. This, in turn, facilitates downstream processes such as the estimation of the NoC to a mixed DNA sample. Peaks classified as stutter within the software can be included optionally or excluded when outputting results and creating EPG reports.

Composite Stutter Positions

Alleles in a profile may be positioned so that multiple different stutter types can all appear in a single peak position. Expected stutter ratio considerations need to be updated for such composite peaks which consist of multiple stutter variants, as well as stutter peaks that may have been affected by spectral pull-up. FaSTR™ DNA includes options to detect and adjust stutter filter and flag ratios for an unrestricted number of stutter types within composite stutter positions.

Known Artefact Detection

To complement stutter detection within a profile, FaSTR™ DNA also allows known artefacts to be configured within the kit. These can be configured as either fixed position artefacts or as artefacts that appear in a position relative to a parent allele (similar to stutter). Peaks appearing in these positions can be filtered and automatically removed, marked for analyst review, or just noted within the peak details.

Estimation of the Number of Contributors to a DNA Profile

The estimation of the NoC to a forensic DNA profile is an important step in the interpretation process. Manual assessment of the NoC by an analyst is a time-consuming task and can be subjective. By default, FaSTR™ DNA is provided with one trained decision tree for NoC estimation of profiles generated using the GlobalFiler™ profiling kit and another for the PowerPlex® Fusion 6C profiling kit. The analyst may also manually assign NoC to a sample or override an existing NoC estimation within the software.

Reports and Exports

FaSTR™ DNA is able to export results in .csv or .text file format and create individual PDF reports for EPGs with a number of customizable settings. An analysis summary PDF report is also able to be generated within FaSTR™ DNA. The new version of FaSTR™ DNA allows default export settings to be saved for an analysis method, including the default export template to use. Notes that can be added to the EPG during analysis or to a snip taken during analysis can also be printed to the EPG report.

Workflows

Since a full project of samples may contain samples from multiple different cases, the new version of FaSTR™ DNA includes the ability to trim or split analysis projects down to analyst-selected samples for more efficient case management.

The review module within FaSTR™ DNA enables comparison of two projects containing the same set of samples and analyzed using the same method to provide a side-by-side comparison of any differences in analysis (including peak label, peak removal/retention, peak classification, and peak size) with the EPG profiles of both projects simultaneously in view. The reviewing analyst can create a final reviewed project and create summary and EPG reports or export results to file or directly to STRmix™ from this module.

A comparison module in FaSTR™ DNA allows sample-to-sample comparison within a project, as well as comparison of samples within the project to a saved database. This can be used to investigate sample-to-sample allele correspondence, as well as correspondence to, for example, a staff elimination database.

In the sample-to-sample comparison run in FaSTR™ DNA, samples are compared pair-wise within a project and matches are generated where the number of alleles in common between two samples exceeds the match threshold configured. The matches are based on counts of peaks matching where minimum values can be set for either the number of matching peaks, or the number of loci with matching peaks to reduce adventitious matches. Matching samples can be inspected side by side in the analysis review module and replicate groups can be created from within the comparison module.

In FaSTR™ DNA’s database comparison feature, each sample within a project is compared against individuals from a database (e.g., a staff elimination database). The allele matching mechanism can be the same as the sample-to-sample matching, but in this feature the database entries (individuals) are the subject of the comparison match. Any project samples that match (above the match threshold settings) to a database individual are listed as matches against that individual.

Analysis to Mixture Interpretation

Within FaSTR™ DNA, once samples are analysed to completion and the NoC assigned, the project can then be exported to STRmix™ for either an automated deconvolution batch or further setup of an interpretation batch with references for a likelihood ratio (LR) calculation. These batch setups are undertaken within the Batch Maker module, available within STRmix™ versions 2.10 and higher. In the new version of FaSTR™ DNA, replicate interpretations can be grouped within FaSTR™ DNA prior to exporting to STRmix™ for interpretation.

The algorithms used for analysis in FaSTR™ DNA are conceptually adapted from the proven approach of OSIRIS analysis software (National Center for Biotechnology Information). The developmental validation of FaSTR™ DNA is published and available[1].

Like its previous iterations, the new version of FaSTR™ DNA was driven through extensive engagement with the end user community to ensure the feature set would be targeted to real world needs and useful for casework operations.

For more information about the new version of FaSTR™ DNA or any of the STRmix team’s other software solutions, visit https://www.strmix.com.

STRmix – A global leader in forensic DNA interpretation, trusted by more than 120 forensic laboratories worldwide.