Zamin Iqbal
@zaminiqbal.bsky.social
Professor of Algorithmic and Microbial Genomics at the University of Bath (UK). Pangenomes, drug resistance (esp TB), data structures for DNA search, plasmid evolution, global microbial surveillance. Open Data, reproducibility
Excited to speak today at UQ, Brisbane. 37 degrees today, and expecting hail and storm later today!
November 27, 2025 at 4:59 AM
Excited to speak today at UQ, Brisbane. 37 degrees today, and expecting hail and storm later today!
Very excited to be in Adelaide to attend #ABACBS2025 . Australia is a powerhouse of microbial genomics, and indeed of bioinformatics, so am very much looking forward to meeting everyone, old friends and new, and speaking tomorrow!
November 24, 2025 at 12:11 AM
Very excited to be in Adelaide to attend #ABACBS2025 . Australia is a powerhouse of microbial genomics, and indeed of bioinformatics, so am very much looking forward to meeting everyone, old friends and new, and speaking tomorrow!
What do you call these kinds of plot? Mind gone blank. (This is from www.nature.com/articles/nat... ).
November 10, 2025 at 8:55 PM
What do you call these kinds of plot? Mind gone blank. (This is from www.nature.com/articles/nat... ).
Wait, Samuel L Jackson was in Jurassic Park!!!??
October 24, 2025 at 7:58 PM
Wait, Samuel L Jackson was in Jurassic Park!!!??
I like this sculpture, spotted in Bristol
October 11, 2025 at 8:52 PM
I like this sculpture, spotted in Bristol
St George's Hall, Bristol. At a Cerys Hafana concert!
ceryshafana.com
ceryshafana.com
October 2, 2025 at 7:29 PM
St George's Hall, Bristol. At a Cerys Hafana concert!
ceryshafana.com
ceryshafana.com
I see chi/rho, and alpha/omega, which all fits the Christian theme. But is that agioc in caps above the columns? What does that mean?
October 2, 2025 at 6:50 PM
I see chi/rho, and alpha/omega, which all fits the Christian theme. But is that agioc in caps above the columns? What does that mean?
We also compared genomes from different batches of the same strain, freeze dried in 1951 & 1998, and mapped reads to both. The genome was fully covered for both- it's conceivable MGE's caused rearrangements that we did not spot, we did not specifically check for that,but the genomes seemed the same
October 1, 2025 at 12:04 PM
We also compared genomes from different batches of the same strain, freeze dried in 1951 & 1998, and mapped reads to both. The genome was fully covered for both- it's conceivable MGE's caused rearrangements that we did not spot, we did not specifically check for that,but the genomes seemed the same
I can only speak for the Murrays, a subset of NCTC, which I think didn't undergo enough freeze-dry cycles to make a v big difference. (Details buried in supplement of our paper)
October 1, 2025 at 12:01 PM
I can only speak for the Murrays, a subset of NCTC, which I think didn't undergo enough freeze-dry cycles to make a v big difference. (Details buried in supplement of our paper)
We found two things. First - a small fraction of the ancestral plasmids evolved into the modern day vectors of antibiotic resistance. These were primarily conjugative (self-transmissible) plasmids. We also see an increase in the numbers of transposons borne.
10/n
10/n
September 25, 2025 at 9:36 PM
We found two things. First - a small fraction of the ancestral plasmids evolved into the modern day vectors of antibiotic resistance. These were primarily conjugative (self-transmissible) plasmids. We also see an increase in the numbers of transposons borne.
10/n
10/n
Our question was - how have plasmids adapted to the era of human industrialisation of antibiotic use ? To answer this, we went back to a culture collection of bacterial isolates from 1917-1954 - the Murray Collection.
6/n
6/n
September 25, 2025 at 9:34 PM
Our question was - how have plasmids adapted to the era of human industrialisation of antibiotic use ? To answer this, we went back to a culture collection of bacterial isolates from 1917-1954 - the Murray Collection.
6/n
6/n
Delighted to see our paper studying the evolution of plasmids over the last 100 years, now out! Years of work by Adrian Cazares, also Nick Thomson @sangerinstitute.bsky.social - this version much improved over the preprint. Final version should be open access, apols.
Thread 1/n
Thread 1/n
September 25, 2025 at 9:29 PM
Delighted to see our paper studying the evolution of plasmids over the last 100 years, now out! Years of work by Adrian Cazares, also Nick Thomson @sangerinstitute.bsky.social - this version much improved over the preprint. Final version should be open access, apols.
Thread 1/n
Thread 1/n
Wei developed a very new approach to indexing
millions of genomes, allowing full alignment. Figure shows: time taken to align 2 genes, as size of the database grows to 1 million
genomes LexicMap (green) is faster (lower on y axis; log scale!!) and uses less RAM (smaller blobs). 4/n
millions of genomes, allowing full alignment. Figure shows: time taken to align 2 genes, as size of the database grows to 1 million
genomes LexicMap (green) is faster (lower on y axis; log scale!!) and uses less RAM (smaller blobs). 4/n
September 10, 2025 at 9:16 AM
Wei developed a very new approach to indexing
millions of genomes, allowing full alignment. Figure shows: time taken to align 2 genes, as size of the database grows to 1 million
genomes LexicMap (green) is faster (lower on y axis; log scale!!) and uses less RAM (smaller blobs). 4/n
millions of genomes, allowing full alignment. Figure shows: time taken to align 2 genes, as size of the database grows to 1 million
genomes LexicMap (green) is faster (lower on y axis; log scale!!) and uses less RAM (smaller blobs). 4/n
There has been huge progress on the standard problem
of mapping to a single reference over the years,
but algorithms for BLAST-ing against a huge database have not
been able to scale with data growth 2/n
of mapping to a single reference over the years,
but algorithms for BLAST-ing against a huge database have not
been able to scale with data growth 2/n
September 10, 2025 at 9:13 AM
There has been huge progress on the standard problem
of mapping to a single reference over the years,
but algorithms for BLAST-ing against a huge database have not
been able to scale with data growth 2/n
of mapping to a single reference over the years,
but algorithms for BLAST-ing against a huge database have not
been able to scale with data growth 2/n
A sad day! what about the crazy machines de l'ile - I hope they remain!
August 1, 2025 at 11:58 AM
A sad day! what about the crazy machines de l'ile - I hope they remain!
Wonderful city - do they still have the amazing slide down the outside of the castle (seen in 2017)?
August 1, 2025 at 11:30 AM
Wonderful city - do they still have the amazing slide down the outside of the castle (seen in 2017)?
Anyone happen to know what these flowers are? Seen in Edinburgh (also seen separate from the horsetails)
July 2, 2025 at 4:37 PM
Anyone happen to know what these flowers are? Seen in Edinburgh (also seen separate from the horsetails)
Farewell #ABPHM25, a great conference! Thanks to the organisers and the great community at the intersection of methods, microbiology and public health! Someone took a picture of my last slide, a message to PhD students. Choose the right data for your question, and don't assume you need ALL the data!
May 23, 2025 at 5:53 PM
Farewell #ABPHM25, a great conference! Thanks to the organisers and the great community at the intersection of methods, microbiology and public health! Someone took a picture of my last slide, a message to PhD students. Choose the right data for your question, and don't assume you need ALL the data!
First id of West Nile Virus in mosquitoes in the UK
May 21, 2025 at 11:02 PM
First id of West Nile Virus in mosquitoes in the UK
@danderson123.bsky.social did a great 2 minute summary of his work on multicopy AMR gene detection,l
#ABPHM25
#ABPHM25
May 21, 2025 at 4:35 PM
@danderson123.bsky.social did a great 2 minute summary of his work on multicopy AMR gene detection,l
#ABPHM25
#ABPHM25
Each point is a gene, size representing the number of samples it is in. Amira systematically improves recall. Our interpretation is that this is quantifying the number of AMR genes that are broken by contig breaks in Flye assemblies. (19/n)
May 19, 2025 at 9:39 AM
Each point is a gene, size representing the number of samples it is in. Amira systematically improves recall. Our interpretation is that this is quantifying the number of AMR genes that are broken by contig breaks in Flye assemblies. (19/n)
Precision at the nucleotide level was good for all tools for the R9.4.1 and R10.4.1 datasets, with mean nucleotide identities of 99.9% and 99.9% respectively for Amira, compared to 98.4%1
and 99.6% for AMRFP Flye, 99.3% and 99.6% for AMRFP Raven, and 99.6% and 99.6% for AMRFP Unicycler.
15/n
and 99.6% for AMRFP Flye, 99.3% and 99.6% for AMRFP Raven, and 99.6% and 99.6% for AMRFP Unicycler.
15/n
May 19, 2025 at 9:36 AM
Precision at the nucleotide level was good for all tools for the R9.4.1 and R10.4.1 datasets, with mean nucleotide identities of 99.9% and 99.9% respectively for Amira, compared to 98.4%1
and 99.6% for AMRFP Flye, 99.3% and 99.6% for AMRFP Raven, and 99.6% and 99.6% for AMRFP Unicycler.
15/n
and 99.6% for AMRFP Flye, 99.3% and 99.6% for AMRFP Raven, and 99.6% and 99.6% for AMRFP Unicycler.
15/n
Then we benchmark primarily against assembly+AMRFinderPlus using either Fly or Raven (this is all nanopore only), but also against (hybrid) Unicycler+AMRFinderPlus. 14/n
May 19, 2025 at 9:35 AM
Then we benchmark primarily against assembly+AMRFinderPlus using either Fly or Raven (this is all nanopore only), but also against (hybrid) Unicycler+AMRFinderPlus. 14/n
So - how does it work? Two types of benchmark, Dan did some nice simulations of progressively more difficult multicopy multi-gene arrays. Recall was systematically better than Flye/AMRFinderPlus across all depths.
11/n
11/n
May 19, 2025 at 9:34 AM
So - how does it work? Two types of benchmark, Dan did some nice simulations of progressively more difficult multicopy multi-gene arrays. Recall was systematically better than Flye/AMRFinderPlus across all depths.
11/n
11/n