humanitarian · health · family: it worked in the lab
the nameswe cannot return
Conflict-zone mass grave identification fails when DNA is too degraded
Problem statement
An estimated 100,000+ unidentified remains from recent conflicts lie in mass graves across Syria, Iraq, Ukraine, the Balkans, and Latin America. Standard STR (short tandem repeat) DNA profiling — the gold standard for forensic identification — fails on heavily degraded skeletal remains exposed to tropical heat, burial chemistry, and relocation. Even when DNA can be extracted, matching requires ante-mortem reference samples from families scattered across multiple countries as diaspora. No field-deployable rapid sequencing solution or scalable diaspora sample collection system exists.
Why this matters
Families of the missing experience ambiguous loss — unable to grieve, resolve legal status, or access inheritance/benefits. International humanitarian law requires states to account for the dead and missing in armed conflict. Identification enables justice processes (war crimes prosecution), family closure, and legal resolution. The ICMP's Bosnia experience showed identification is possible but took 20+ years with centralized, well-funded labs — a timeline that current conflicts (Syria, Ukraine, Sudan) cannot repeat.
What’s been tried and why it hasn’t worked
The ICRC's AM/PM Database (14 countries since 2007) manages ante-mortem/post-mortem data but relies on STR profiling that fails on degraded samples. The ICMP pioneered mass DNA-led identification in the former Yugoslavia (more than 70% of 40,000 missing accounted for) but required centralized lab infrastructure and decades. Next-generation sequencing (NGS) and SNP panels can recover information from highly degraded material but cost $50–$150/sample, require expensive equipment, trained molecular biologists, and weeks of processing. Oxford Nanopore MinION offers portable sequencing but has not been adapted for forensic bone/tooth samples in austere conditions. Ante-mortem data collection from displaced families is itself a bottleneck — families may be suspicious of authorities, culturally resistant, or simply unreachable.
What would unlock progress
Field-deployable DNA extraction and targeted sequencing platforms adapted for severely degraded bone and tooth samples in austere conditions (limited power, high temperature, no clean lab). Diaspora-accessible family reference sample collection systems allowing remote DNA submission with chain-of-custody integrity. ML reconciliation algorithms that match partial, low-quality post-mortem profiles against incomplete family reference databases.
Entry points for student teams
A team could adapt a portable sequencing workflow (MinION-based) for degraded bone samples, testing with archived reference materials to characterize the degradation threshold for usable profiles. Alternatively, a team could design a secure, remote sample collection kit and chain-of-custody protocol for diaspora families. Molecular biology, forensic science, and humanitarian engineering skills apply.
Genes
Structural cousins: stuck the same way, in other fields
Sources
ICRC, Missing People, DNA Analysis and Identification of Human Remains: A Guide to Best Practice in Armed Conflicts and Other Situations of Armed Violence, second edition 2009; ICRC and EPFL Humanitarian Tech Hub, accessed 2026-02-24 go to source ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
The ICRC-EPFL Humanitarian Tech Hub (launched 2016) is actively developing technology solutions for this problem space. The Syria conflict alone has an estimated 130,000+ missing and disappeared persons. Ukraine's prosecutor general has documented thousands of civilian graves needing forensic examination. The forensic identification pipeline has three bottlenecks (sample extraction, sequencing, matching) — all three need simultaneous improvement for the pipeline to scale.
Reconciliation 2026-09-29: A site review read this write-up against its sources. The cited guide was fetched on 2026-09-29 from the address in the Source line, and its English PDF (52 pages) was downloaded from the ICRC shop at https://shop.icrc.org/download/ebook?sku=4010/002-ebook and read in full. The cover says "Second edition 2009". The Humanitarian Tech Hub has no address in the Source line, so two pages about it were fetched the same day: the ICRC news release at https://www.icrc.org/en/document/icrc-and-epfl-launch-humanitarian-tech-hub and the EPFL news item at https://actu.epfl.ch/news/the-humanitarian-tech-hub/. (1) The Source line gave a shortened title and no year. It now carries the full title and the edition copied from the PDF cover. (2) The text said the Tech Hub was "launched 2023". The ICRC news release is dated 8 March 2016 and the EPFL page says the hub was launched in 2016. The year now reads 2016. (3) The text said the ICMP identified "~70% of 40,000 missing" in Bosnia. The guide does not name the ICMP or give these figures; it says only that one of the two largest DNA-led programmes helped identify thousands of people killed in the Balkans between 1991 and 2000. The ICMP's own Western Balkans page (https://icmp.int/what-we-do/geographic-programs/western-balkans/, fetched 2026-09-29) says more than 70 percent of the 40,000 people who went missing in the conflicts in the former Yugoslavia have been accounted for. Its Bosnia and Herzegovina page gives a different figure for that country, approximately 75 percent. The sentence now names the former Yugoslavia and says "accounted for". This page is not one of the write-up's cited sources. Confirmed against the guide and left as written: STR analysis does not always succeed on degraded remains; warm and humid conditions destroy DNA faster; SNPs may be the only marker that can be analysed when DNA is highly degraded; matching needs reference samples from relatives; large programmes take several years; chain of custody matters for sample collection; the ICRC has a database for ante-mortem and post-mortem data. Confirmed against a page that is not a cited source and left as written: more than 130,000 people missing from the Syrian conflict (https://icmp.int/the-missing/where-are-the-missing/syria/). NOT checked, because they are not in the guide or in the two Tech Hub pages and no cited source for them could be found: the statement that the Tech Hub is actively developing technology for this problem (the ICRC release lists fields such as energy, water, logistics and biomedical technologies, names a prosthetic foot as the first product, and does not mention forensic identification; the hub's later work was not read); the estimate of 100,000+ unidentified remains and the list of countries; the statement that no field-deployable sequencing or diaspora sample collection system exists; the ICMP timeline of 20+ years; "14 countries since 2007" for the ICRC database, and the statement that the database relies on STR profiling; the $50–$150 per sample cost of newer sequencing methods; the statements about Oxford Nanopore MinION; the statement about Ukraine's prosecutor general; and the three-bottleneck summary. The guide dates from 2009, so it cannot be the source for anything about Syria, Ukraine, Sudan or portable sequencers. These are left as written. Genes untouched.