BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260421T090513Z
LOCATION:Bldg. 6 - Room 103
DTSTART;TZID=Europe/Stockholm:20260629T163000
DTEND;TZID=Europe/Stockholm:20260629T170000
UID:submissions.pasc-conference.org_PASC26_sess158_msa170@linklings.com
SUMMARY:Pangenome Alignment at Scale: HPC Challenges and Acceleration Stra
 tegies
DESCRIPTION:Santiago Marco-Sola (Barcelona Supercomputing Center)\n\nPange
 nome graph representations are increasingly replacing single linear refere
 nces, fundamentally changing how genomic analyses are performed at the pop
 ulation scale. However, this transition introduces significant computation
 al challenges, particularly in sequence alignment against graph-based refe
 rences, where complexity, memory usage, and scalability become limiting fa
 ctors on current computing systems. As sequencing projects continue to exp
 and in size and scope, the ability to efficiently align reads against larg
 e pangenome graphs is becoming a critical constraint for the viability of 
 population-scale genomics. \n \nIn this talk, we will discuss the HPC chal
 lenges posed by pangenome alignment, drawing from our recent work on seque
 nce-to-sequence and sequence-to-graph alignment algorithms. We will presen
 t how algorithmic and data-structure design choices interact with modern H
 PC architectures, exploring the trade-offs between computational cost, res
 ource usage, and scalability. Our work demonstrates that GPU-based and het
 erogeneous hardware acceleration are paramount for overcoming current perf
 ormance limitations and examines how these accelerators can be leveraged t
 o scale pangenome alignment on modern HPC platforms. Furthermore, we discu
 ss open challenges and future directions for scalable pangenome alignment 
 on HPC systems, arguing that contributions in this area require tight soft
 ware-hardware co-design across algorithms and hardware accelerators to cop
 e with the ever-growing scale of population-level pangenomic datasets.\n\n
 Domain: Engineering, Life Sciences, Computational Methods and Applied Math
 ematics\n\nSession Chairs: Gagandeep Singh (AMD), Sriranjani Sitaraman (AM
 D), Bertil Schmidt (Johannes Gutenberg University Mainz), Kristof Denolf (
 AMD), Mittul Singh (AMD), and Yijie Xu (AMD)\n\n
END:VEVENT
END:VCALENDAR