Heinrich Ferreira
LightNews LightNews
banner
heinrichferreira.bsky.social
Heinrich Ferreira
@heinrichferreira.bsky.social
230 followers 120 following 18 posts
Sydney, Australia. Lover of non-fiction books on biology and curious about bioinformatics.
Created the Visual Genome Browser, which represents genomic data in 2D.
https://splicejunction.blogspot.com/2015/09/spliced-genes-natures-model-hobby-kit.html
Posts Replies Media Videos
heinrichferreira.bsky.social
Finding which bacteriophages a bacteria has CRISPR resistance against. I have been using the #VisualGenomeBrowser to do a #blast search through the enormous 32043 phage #fasta database from millardlab.org. Read the alt text for details.
#crispr #phage #cas9 #protospacer
What you see here is the CRISPR locus of the bacteria: Vagococcus penaei strain CD276:998300-1000035 I search for CRISPR repeats of between 66 and 73 bases in length in a bacterial genome and then I try to match each of the spacer sequences found between the repeats to the phage db. The spacers contain pieces of DNA taken from past infections. More recent infections are inserted closer to the start of the locus. By building up a list of bacteria and the phage locations from which the spacers were taken it is possible to build up a tree of bacteria and the phage that infected them, the phage and which multiple bacteria each of them have infected and have spacers, also which bacteria have common repeat motifs due to having similar CRISPR systems. Lastly it is also possible to find bacteria which either have spacers from the same phage because they both got infected by the same phage OR because they descend from the same ancestor which got infected by the same phage and survived to tell the tale. Usually the latter would have the spacers from different places in the same phage genome and that is how you tell the difference. This is the CRISPR locus from Streptococcus thermophilus strain KLDS 3_1003:121258-122018 (Bases=761) (Strand=POS+) You can see different spacers from multiple different phage that got integrated into the bacterial genome between the repeats. It is possible to use BLAST to know exactly from what part of each phage the spacer was taken. The spacer gets integrated without the PAM sequence (protospacer adjacent motif eg. NGG which is only attached to the spacer when it is still in the phage) since that prevents the bacteria from targeting the spacer when it is part of its own genome. This is the CRISPR locus of the Streptococcus pyogenes strain 7F7:751066-751629 (Bases=564) (Strand=POS+) Here one can see that there are spacers from the same JAVAN 447 phage but from different locations in that phage.
January 8, 2025 at 8:42 PM
heinrichferreira.bsky.social
One of the most universal protein machines across all domains of life. Here is the ATP Synthase of cattle (PDB: 5ARA). Image on the left shows that all subunits except the F0 Beta subunit (which is found on CHRM) are coded on the nuclear genome.
#PDB #bioinformatics #atpsynthase
What you can see here is a screenshot of ChimeraX showing the protein databank PDB: 5ARA of an ATP Synthase multi-subunit protein machine used for producing ATP from the flow of Hydrogen ions down the concentration gradient from the high concentration in the inter membrane space into the lower concentration mitochondrial matrix. All subunits except the F) Beta subunits that are used in the mitochondrial membrane are coded for on nuclear chromosomes. The F0 component ATP6,ATP8 are coded for on the mitochondrial circular genome. All others have been relocated into the nuclear genome as it has better error correction, break repair and is less exposed to oxidative damage. The F0 subunit encoding was kept on the mitochondrial genome as it allows the mitochondria to react quicker to changes in energy requirements by being able to maintain local protein production inside the mitochondria. All other subunits' proteins have to have mitochondrial targeting signals to allow them to be transported to the mitochondrial. This allows the mitochondria to quickly shut off ATP synthesis by halting the synthesis of the F0 subunit. Since all components are needed for it to function it can switch off operation by simply removing one crucial subunit. What you see here in the Chimera-X software package is the 5ARA cattle (Bovine) ATP Synthase obtained by X-Ray crystallography and the proteins to the left and right are predicted protein structures that I created by passing the 22 amino acid sequences to Alphafold3. (11 distinct sequences with some components such as AT5G1 repeating). I basically just threw the FASTA sequences into a JSON file and uploaded it. Alphafold 3 was able to accurately combine the predicted folded proteins into the correct multi subunit enzyme structure. Amazing! Alphafold 3 can be found at https://golgi.sandbox.google.com/ These are the amino acid subunit sequences I uploaded to Alphafold 3 to obtain the multi subunit predicted structure.
January 2, 2025 at 10:39 PM
heinrichferreira.bsky.social
Mapped reads from a CRAM file by using SAMTOOLS in the background to interactively decode the DNA reads overlapping with the telomeric region of Chromosome 1 from the HG38 human reference genome.
Coming functionality for Visual Genome Browser
#bioinformatics #genomics #telomeres #cram #genomebrowser
Display of mapped reads from a CRAM file using SAMTOOLS to interactively decode the DNA reads overlapping with the telomeric region of Chromosome 1 from the HG38 human reference genome.
January 2, 2025 at 11:08 AM
heinrichferreira.bsky.social
Some of the different ways to differently visualise the satellite repeats in the centromeric region of human chromosome 1. Taken from the Visual genome Browser. The first image is 1869 bases per horizontal line, the second at 340 bases.
#genomics #centromeric #dna #bioinformatics #visualization
Some 2-dimensional patterns obtained by displaying the satellite repeats in the centromere region of human chromosome 1 between coordinates 124851077-124851262 cantered at at 124931537 with each horizontal line of bases offset at 1869 bases (genome data taken from the HS1 telomere-to-telomere reference sequence. The same centromeric repeats but this time shown at an offset of 340 =2x170 bases centred around coordinate chr1:124621327 (Coincidentally the number of bases that it takes to wrap around the CENP1 histone proteins equates to about 170 bases including the linker sequence) This is the Dot-plot view of the same satellite repeat where the centromere region is compared with itself showing how the same repeats occur across the chromosome centromeric region These RNA secondary structures for both the sense and antisense strands were calculated on the 340 base repeats in the centromeric satellite DNA on chromosome 1.
January 2, 2025 at 10:56 AM
heinrichferreira.bsky.social
My unique rearrangement of the genetics code as a Hamming code table where a single nucleotide change in any codon letter most frequently results in a silent or conservative amino acid change with similar properties.
#genomics
#computationalbiology
#bioinformatics
Genetic code codons ordered by Hamming code, representing a "torus" wrapping back on itself horizontally and vertically. The intention is to depict how the generic code is robust against single nucleotide mutations but also allows the code to be implemented with fewer than 61 tRNA by allowing the second and third bases to be able to bind less strictly to the mRNA codons.
January 1, 2025 at 5:23 AM
heinrichferreira.bsky.social
I have just finished another remarkable book by Nick Lane. I just love how he can make complex subjects easy to understand. I really recommend the book: Oxygen by Nick Lane on Audible. www.audible.com.au/pd/149454660...
December 31, 2024 at 11:56 AM