United States v. Jones, 965 F.3d 149

Dean Jones was indicted on narcotics (conspiracy; firearm) and separate Hobbs Act robbery/conspiracy and related firearm counts; trials were bifurcated and he was convicted on the narcotics conspiracy and on the robbery/Hobbs Act and firearm counts and sentenced to 312 months.
The robbery (Dec. 21, 2012) involved a masked, gloved gunman who shot a victim and discarded a hat and a blue latex glove; the hat contained a single-source DNA match to Jones; the glove contained a complex mixed DNA sample.
New York City OCME analyzed the glove mixture using its internally developed Forensic Statistical Tool (FST), which reports likelihood ratios (LRs); FST produced an LR of 1,340 ("very strong" support) that Jones was a contributor.
A five-day Daubert hearing examined FST’s development, validation (including ~500,000 noncontributor comparisons), treatment of drop-in/drop-out, the post-validation allele-cap modification, and OCME’s accreditation; OCME reported an overall false-positive rate of 0.03% and 0.0009% for LRs >1,000.
The district court admitted the Glove DNA/FST testimony under Rule 702 and Daubert; Jones was convicted at trial on the robbery charges. On the narcotics trial, Jones challenges refusal to give a multiple-conspiracy charge and seeks a new trial based on post-trial impeachment evidence about a cooperating witness.

Issue	Plaintiff's Argument	Defendant's Argument	Held
Admissibility of OCME's FST-based Glove DNA under Rule 702/Daubert	FST is validated, peer-reviewed/approved by NYS committees, has very low false-positive rates, and uses accepted LR methodology; admissible	FST is unreliable: uses preset drop-out rates tied to quant (which has ~30% error), allele-cap modification, unique method not used elsewhere; challenges go to weight not admissibility	District court did not abuse discretion admitting FST; Daubert factors support reliability; any error would be harmless given other evidence
Multiple-conspiracy jury instruction (narcotics trial)	Single conspiracy was charged and proved; a multiple-conspiracy instruction was unnecessary	Requested instruction was necessary to prevent jury attributing other co-conspirators' conduct/quantities to Jones	Court permissibly refused the requested instruction as confusing and unnecessary; given Jones was the sole defendant at trial, no prejudice shown
New trial based on newly discovered impeachment evidence about cooperating witness Christopher	Newly revealed misconduct (contraband smuggling) would materially impeach Christopher and could lead to acquittal	Government: impeachment evidence is cumulative; Christopher’s credibility had already been attacked at trial with extensive prior impeachment material	Denial of new trial affirmed; evidence was cumulative, would not probably produce acquittal; standard for Rule 33 relief not met

Daubert v. Merrell Dow Pharm., 509 U.S. 579 (trial-court gatekeeping inquiry under Rule 702)
Kumho Tire Co. v. Carmichael, 526 U.S. 137 (trial court's broad discretion in assessing expert reliability)
Amorgianos v. Nat'l R.R. Passenger Corp., 303 F.3d 256 (2d Cir. 2002) (factors for evaluating admissibility of expert testimony under Rule 702)
United States v. Parkes, 497 F.3d 220 (2d Cir. 2007) (standard for new trial based on newly discovered evidence)
United States v. Maldonado-Rivera, 922 F.2d 934 (2d Cir. 1990) (multiple-conspiracy instruction principles)
United States v. Spencer, 4 F.3d 115 (2d Cir. 1993) (new-evidence/new-trial precedents on impeachment evidence)
United States v. McGinn, 787 F.3d 116 (2d Cir. 2015) (harmlessness review for evidentiary error)