LACTBL1 Notes

OverviewWhat can be confirmed from databases, literature and the PS2 analysis as of 2026-10-05

LACTBL1 encodes a 579-amino-acid protein built mostly on a beta-lactamase-like fold. Its RNA is highest in the retina, and single-cell data place it in retinal bipolar cells. Orthologs exist in every major vertebrate group, and the most distant one-to-one ortholog is in the tunicate, so the gene is at least 600 million years old. The only paralog in humans that can be recognized from sequence is the mitochondrial serine protease LACTB. The two share less than 20% identity and split before cnidarians and bilaterians did. Its rate of evolution is middling: about 3 times that of cytochrome c and half that of fibrinogen alpha. The most conserved residues cluster in the beta-lactamase catalytic core. No one has tested the activity of this protein yet.

Official name: lactamase beta like 1, no aliases Record
HGNC
Location: chr1:22,953,043–22,972,866 (GRCh38); 1p36.12, minus strand, spans 19,824 bp Record NCBI Gene
Transcript: Only one: NM_001289974.2 (MANE Select), 8 exons, no alternative splicing Record
mRNA: 1,950 nt; 5′UTR 1–210, CDS 211–1950; the record ends at the stop codon, with no 3′UTR Record
Protein: 579 aa, 63,506 Da, pI 9.00 Record ProtParam
Domain: AmpC region aa 70–410 (NP record); Pfam PF00144: NCBI gives 87–411, InterPro gives 87–435 Record
Expression: HPA: retina enriched, bipolar cells at single-cell level; GTEx near 0 in all tissues (GTEx has no retina) Record
Homologs: NCBI lists 553 species (mostly amniotes); Ensembl adds ray-finned fish and the tunicate. Among invertebrates, the tunicate has a one-to-one ortholog, the sea anemone has 8 co-orthologs, and the sponge is unclear Analysis Evolution
Disease: No OMIM entry, no Mendelian disease; only weak GWAS associations (refractive error, a suggestive signal for premature ejaculation) Literature

Data portal66 links, each pointing straight to a page for this gene, transcript or protein

Filter

NCBI core records

NCBI search and analysis

Genome and nomenclature

Protein and structure

Expression

Variants, disease and mouse

Literature and patents

Homologs (for Q3)

Course tools

Course scheduleAll 29 classes and every due date for the semester, in Chicago time. Dates checked against Canvas and the course schedule on 9/30. Open a past class to see my notes: what was covered and what I did with LACTBL1. Slides are on Canvas (UMN login required)

Today9/30 WedWeek 4 of 15

Done7 / 2922 left

Next class10/05 Mon#8 Multiple sequence alignment

Next due10/01 23:59PD1 PS1 draft · tomorrow

Unmarked classes are in the classroom · ◆ Due · ⚠ Canvas and the schedule disagree

Round 1 · Gene identity, sequence, alignment

9/9 – 10/4 · PS1 due: draft 5 · peer review 3 · final 35

9/09Wed#1What this course is, and how to pick a gene "nobody has studied"The whole semester is spent on one human gene of unknown function, ending in a report and a Wikipedia article. Week one is about learning how to pick that gene.Due todayCS1Pre-course survey23:59 · extra credit
What was covered

Bioinformatic analysis means studying DNA, RNA and protein with free online databases and tools: finding sequences, translating, aligning, checking expression, predicting structure and function. It is faster than wet lab work and costs only time.

No exams. The grade comes from 4 Problem Sets, 4 group presentations, a final report, one Wikipedia article, plus peer review.

How to tell "unknown function": no Wikipedia page (or just a few lines), and no PubMed paper with the gene name or an alias in the title.

How to search: in NCBI, use "Homo sapiens"[organism] plus terms like unknown function, hypothetical, TMEM#, FAM#, C#orf#. Recheck every candidate under all its aliases.
What I did

Started screening candidate genes by these criteria and kept the process in a search log.
Slides: Slides 1-1, 1-2 · Task this week: pick 3 candidate genes and enter them in the class spreadsheet
9/14Mon#2Finding bioinformatic information: where to look for a candidate geneTake a candidate gene through a fixed checklist to decide whether it really is unstudied, and collect everything you can find along the way.
What was covered

Order of checks: Wikipedia → NCBI Gene (look at the PubMed papers linked from the page) → search again under each alias.

Demo with C1orf50: a direct PubMed search gives 0 papers, but the NCBI Gene page links several. So check both.

What each database gives: Nucleotide for mRNA length, exons, UTRs, polyA signal; Protein for amino acid count and molecular weight; GeneCards for aliases, domains and expression in one place. Also check Google Scholar, OMIM, AceView, Google Books and the US patent office.

Search habits: know what question you are answering first, list all synonyms, judge quality as you go, and know when to stop. Use ⌘F for keywords in long papers and keep screenshots as evidence.
What I did

Chose LACTBL1. The other two candidates were RNF224 and C7orf33.

Checked it: no Wikipedia page, never in a PubMed title, and none of the papers linked from NCBI Gene are specifically about it.

Put the IDs, location, exons, domains and expression into a gene profile.
Slides: Slides 2-1 · Maps to: PS1 Q1 (text search), Q4 (Wikipedia screenshot)
9/16Wed#3Judging information quality; translating mRNA into proteinFirst half: how to evaluate what you find. Second half: conceptual translation, i.e. reading the mRNA three bases at a time into amino acids, laying them out aligned with the sequence, and annotating.Due todayCQ1ChimeIn23:59 · 2 pts
What was covered

Good information is concise, complete (say what was ruled out), accurate (sourced), relevant, ordered by importance, and current. Write up the search so someone else could redo it.

In-class exercise: score past example assignments against these criteria, item by item.

Conceptual translation: use SIXFRAME to translate all six reading frames of the mRNA, pick the one that reads from ATG through to a stop codon, and delete the amino acids before the start and after the stop.

Annotation tools: font color, circles, translucent shading, vertical lines at exon junctions, margin notes, brackets around domains.
What I did

Did the conceptual translation of LACTBL1 from NM_001289974.2: frame +1, from the ATG at nt 211 to the TGA at nt 1948–1950, giving 579 amino acids, identical to NP_001276903.1.

Marked the start and stop codons, 7 exon junctions and the AmpC region. Redid the layout on 9/28: no amino acids under the 5′UTR, both bases and amino acids numbered, uORF and Kozak added.
Slides: Slides 3-1, 2-2 · Maps to: PS1 Q2 · Homework: Wikipedia training, alignments with the chicken and mouse homologs
9/21Mon#4Global sequence alignmentAlign the human protein end to end with the same protein from other species and see which positions never changed. Unchanged positions are usually the most important.Zoom · recorded
What was covered

Similarity and homology are different things: similarity is a percentage; homology means two sequences share a common ancestor, and it is yes or no. Alignment asks how similar is similar enough to infer homology.

Alignment assumes the sequences changed step by step from a common ancestor: bases or amino acids were substituted, inserted or deleted.

Conserved positions tend to matter most for structure and function. But in very close species (e.g. rat and mouse) many positions simply have not had time to change, so more distant comparisons tell you more.

Homology does not mean the same function: one enzyme evolved into a structural protein in the eye lens.

A substitution matrix scores each substitution: identical rare amino acids score highest, chemically very different swaps are penalized most.

Global alignment suits single-domain proteins that have not diverged much; use local alignment for highly diverged, multi-domain ones.

Hands-on: find mouse, chicken, zebrafish and fruit fly homologs in NCBI Protein and align them to the human sequence with EMBOSS Needle.
What I did

Ran Needle (EBLOSUM62) on human LACTBL1 (NP_001276903.1) vs mouse Lactbl1 (NP_001230191.1): identity 81.1%, similarity 86.4%, gaps 0.3%.

The most different stretch is aa 186–200, where mouse has an insertion. The S-x-x-K motif at S142–K145 is also present in mouse.
Slides: Slides 4-1 · Maps to: PS1 Q3 (chicken, zebrafish and most distant species not done yet)
9/23Wed#5Dot plots: finding repeats within a sequencePlot a sequence against itself (or another sequence). Diagonal lines are similar segments; lines parallel to the main diagonal mean the sequence contains repeats.Due todayWT1Wikipedia training 123:59 · extra creditZoom · recorded
What was covered

Local alignment only looks for short stretches of similarity, which suits modular proteins. Many proteins are built from several kinds of domains (examples: coagulation factor XII and tPA), and these modules spread through internal gene duplication and exon shuffling.

How to draw one: one sequence across, one down; set a window (how many residues compared at a time) and a threshold (minimum matches to plot a dot). A looser threshold gives more dots and more noise.

Reading it: in a self comparison the main diagonal is always there and means nothing; parallel lines next to it mean internal repeats.

Tools: Dotlet for protein, DNAdot for nucleic acid.
What I did

LACTBL1 against itself (Dotlet, BLOSUM62, window 15): no parallel lines off the main diagonal, so no internal repeats.

A few scattered short dots are proline-rich segments that happen to look alike (e.g. aa 62–76 vs 447–461), not repeats.
Slides: Slides 4-2 · Maps to: the dot plot in PS1 Q3 · Same week: 4 Wikipedia training modules
9/24to 9/26Wikipedia training 2–4WT29/24 23:59 · extra creditWT39/25 23:59 · extra creditWT49/26 23:59 · extra credit
9/28Mon#6BLAST: searching a whole database for similar sequencesBLAST is local alignment plus database search: it reports only the best-matching stretches, ranked by score, ignoring names. This class covered setting parameters, reading results and saving hits as FASTA.Due todayCQ2ChimeIn23:59 · 2 pts
What was covered

Five kinds of BLAST: blastn, blastp, blastx, tblastn, tblastx. blastp is the first choice for finding homologs: the code is degenerate and DNA changes faster than amino acids, so distant relatives that are unrecognizable at the DNA level can still be recognized at the protein level.

Settings: database refseq_protein. Organism set to Homo sapiens alone finds paralogs; human plus the exclude box finds orthologs. The web page now defaults to ClusteredNR, which merges sequences over 90% identical into one entry and makes it easy to miss homologs, so switch it back.

Reading results: hits are ranked by score, not identity. Read identity together with Query cover: in the slides the orca hit has 81% identity but covers only 69%, so it ranks below a bat hit with 73% identity.

E value: how many hits this good you would expect by chance in a database this large. Smaller is more trustworthy, and the same alignment gets a larger E value in a larger database.

XP_ accessions and entries labeled PREDICTED are computer-predicted sequences. They are usable, but check for misjoins and missing ends.

Twilight zone (not covered separately in the slides; added from Rost 1999): at 20–35% identity, the number alone cannot tell you whether two sequences are homologous. The instructor's rule is to stop once identity drops below 25%.
What I did

Collected 20 ortholog proteins, 21 with human, saved as FASTA: 3 mammals, 2 birds, 3 reptiles, 2 amphibians, 2 lobe-finned fish, 2 ray-finned fish, 2 cartilaginous fish, and 1 each for lamprey, sea anemone, sponge and tunicate. Human first, the rest sorted by identity from high to low; the lowest is 31.8%.

Checked the 4 most distant ones (lamprey, sea anemone, sponge, tunicate) separately: alignment scores against human LACTBL1 are 546–871, against the paralog LACTB only 54–149, so they belong to the LACTBL1 branch.

The paralog file is not done yet. LACTB and LACTBL1 share about 18–23% full-length identity, right in the twilight zone, so the reasoning has to be written out clearly.
Slides: Slides 4-3 · Maps to: PS1 Q3 (find the most distant ortholog), PS2 Q1 · Homework: FASTA of paralogs and 20 ortholog proteins
9/30Wed#7BLAT: extending incomplete sequences at both endsSome mRNA or protein records are missing their ends. First use BLAST to find longer sequences in other species and in ESTs, then use BLAT to check them against the genome and fill in the missing piece.Workshop
What was covered

Goal: extend the mRNA in the 5′ direction (same idea for 3′); the same approach can fill in a missing N or C terminus of a protein.

Method: BLAST the 5′UTR plus the first 5 codons, put the 5′UTRs of primate orthologs into one FASTA side by side, and see which extends furthest and which are cut short. The instructor demoed with SARAF and gained 10 bases from another sequence.

Then search the EST database (huge numbers of short, rough cDNA fragments), which added another 76 bp. Plus/Plus means no reverse complement is needed. With Plus/Minus, the apparent extra 428 bp is an artifact; the real gain is at most 120 bp, and it has to be confirmed by BLAT against the human genome.

Once the 5′ end is extended, check the Kozak sequence (CRNCATGR); it also roughly shows where transcription starts.

Protein example: the pangolin TMEM223 sequence only lines up from human position 34, so its N terminus is missing. BLAT of the human sequence to the pangolin genome hit a sequencing gap, so the only options are another species or another database. The instructor kept this slide on purpose as an example of "BLAT found nothing".
What I did

BLATed all of NM_001289974.2 to hg38: all 1,950 nt matched (100%), chr1:22,953,043–22,972,866, minus strand, spanning 19,824 bp.

Listed the coordinates of the 8 exons and the lengths of the 7 introns (340–6,879 bp): exons 1,950 + introns 17,874, exactly 19,824. Exon 1 is followed by GT, consistent with the GT-AG rule.

Revised the English PS1 to version 4 and brought it to the Workshop.

Next step: this mRNA record has no 3′UTR (see Still to check), so I can use this class's method to look for ESTs in the 3′ direction.
Slides: Slides 4-4 · Maps to: gene structure in PS1 Q1 · Coming up: PD1 Thu, PR1 Sat noon, PS1 Sun
10/01ThuPD1PS1 draft23:59 · 5 pts✓ Submitted, and sent to peer review partner Viet (Pair 3)Due next
10/03SatPR1PS1 peer review12:00 · 3 pts⚠ Canvas says 12:01, the schedule says noon; submit before 12:00✓ Submitted on Canvas 10/3: wrote a review of Viet's LOC124900986. His feedback for me: sentences too long; asked where the poly(A) is
10/04SunPS1Final23:59 · 35 ptsWord or PDF; name and gene name in the header; do not copy the questions. Final v7 revised per Viet's peer review: shortened sentences throughout, rewrote 1.1, added a poly(A) note (the record stops at the TGA and has no 3′UTR). On 10/5 also checked Google Scholar and USPTO

Round 2 · Evolution and expression

10/5 – 10/25 · PS2 due: draft 5 · peer review 3 · final 35; plus the PS1 meeting and Symposium 1

PS2 progress (10/4, draft v33): 12 of 15 items are in the draft. The AI statement and full-draft check are half done; the expression question waits for the 10/19 and 10/21 classes. The MSA, trees and rates were done ahead of class; the methods still need checking against the 10/7, 10/12 and 10/14 slides. Results are under Evolution.

10/05Mon#8Multiple sequence alignment: choosing which sequences to compareHow to make a multiple sequence alignment; how to choose sequences to compareThis site: Evolution · Slides posted: Slides 5-1 · Already ran three MSAs with FAMSA (17 strict orthologs, 8 distant homologs, 12 paralogs); compare with Clustal Omega after classNext class
10/05to 10/09Book a meeting with the instructor on PS1 feedback15–20 min, bring a printed copy; submit CR1 on the day of the meeting
10/07Wed#9Editing and formatting an MSA; reading conservation and chemistryHow to edit and format a multiple sequence alignment for conservation and chemistryThis site: Evolution (the 29 most conserved residues) · Update coloring and consensus line to match the slides
10/09FriCR1PS1 meeting feedback checklist23:59 · 2 pts⚠ Syllabus says submit on the day of the meeting
10/12Mon#10Symposium 1 presentations; reconstructing a gene's evolutionary originsSymposium 1; How to reconstruct a gene’s evolutionary originsThis site: Evolution (origins, two unrooted trees) · Check tree-building method against the slidesDue todaySP1Oral presentation12:30 · 5 ptsSR1Presentation peer review23:59 · 3 ptsCQ3ChimeIn23:59 · 2 ptsPresentation
10/14Wed#11Rate of evolution; when paralogs were duplicatedHow to determine rate of evolution; how to determine the date of gene duplication for paralogsThis site: Evolution (rate, LACTB duplication date) · Check the formula for m and the fitting method against the slides
10/19Mon#12Expression level and pattern across human tissuesHow to determine a gene's full expression level and pattern in human tissuesThis site: Expression (HPA: retinal bipolar cells)
10/21Wed#13Expression under different physiological conditionsHow to determine a gene's full expression pattern under different physiological conditions; WorkshopThis site: ExpressionWorkshop
10/22ThuPD2PS2 draft23:59 · 5 pts
10/24SatPR2PS2 peer review12:00 · 3 pts
10/25SunPS2Final23:59 · 35 pts

Round 3 · Regulation, RNA, protein primary structure

10/26 – 11/15 · PS3 due: draft 5 · peer review 3 · final 35; plus the PS2 meeting and Symposium 2

10/26Mon#14Finding promoters, regulatory sequences and eQTLsHow to find promoters and other regulatory sequences including a gene’s eQTLsThis site: Gene and locus, upstream SNP rs10158878Due todayCQ4ChimeIn23:59 · 2 pts
10/28Wed#15Promoters, regulatory sequences and eQTLs (continued)How to find promoters and other regulatory sequences including a gene’s eQTLsThis site: Gene and locus
10/30FriCR2PS2 meeting feedback checklist23:59 · 2 ptsBook the meeting first
11/02Mon#16Symposium 2 presentationsDue todaySP2Oral presentation12:30 · 5 ptsSR2Presentation peer review23:59 · 3 ptsPresentation
11/04Wed#17Predicting RNA structure; finding miRNA and RNA-binding protein regulatory sitesHow to predict RNA structure and sites of regulation by miRNA and RNABPsThis site: Transcript (uORF in the 5′UTR)
11/09Mon#18Analyzing a protein's primary sequence; drawing a protein diagramHow to analyze a protein's primary sequence; how to make a protein diagramThis site: ProteinDue todayCQ5ChimeIn23:59 · 2 ptsZoom
11/11Wed#19Predicting post-translational modificationsHow to predict post-translational modification; WorkshopThis site: Protein (4 candidate N-glycosylation sites)ZoomWorkshop
11/12ThuPD3PS3 draft23:59 · 5 pts
11/14SatPR3PS3 peer review12:00 · 3 pts
11/15SunPS3Final23:59 · 35 pts

Round 4 · Localization, structure, variants, interactions

11/16 – 12/6 · PS4 due: draft 5 · peer review 3 · final 35; plus the PS3 meeting and Symposium 3

11/16Mon#20Predicting subcellular and tissue-specific localizationHow to predict protein subcellular and tissue-specific localizationThis site: Expression (HPA subcellular), Protein (N-terminal hydrophobic segment)
11/18Wed#21Predicting, exploring and annotating secondary structureHow to predict, explore & annotate 2° structuresThis site: Protein
11/20FriCR3PS3 meeting feedback checklist23:59 · 2 pts⚠ No date set in Canvas; using the schedule
11/23Mon#22Symposium 3 presentationsDue todaySP3Oral presentation12:30 · 5 ptsSR3Presentation peer review23:59 · 3 ptsCQ6ChimeIn23:59 · 2 ptsPresentation
11/25Wed#23Predicting and annotating tertiary structure with local and global approachesHow to predict, explore & annotate 3° structures using local and global approachesThis site: Protein (AlphaFold A8MY62) · Thanksgiving weekRecorded
11/30Mon#24Mapping known mutations onto a gene; estimating the health impact of variantsHow to map known mutations to a gene; how to estimate health impact of variantsThis site: Literature and associations (GWAS)
12/02Wed#25Finding a protein's functional partners and interpreting the interactionsHow to identify a protein's functional partners and make sense of the interactions; WorkshopThis site: Literature and associations (PRC1, ZC3HAV1)Due todayPD4PS4 draft23:59 · 5 pts⚠ Canvas says 12/3, the schedule says Wed midnight; submit by the earlier oneWorkshop
12/05SatPR4PS4 peer review12:00 · 3 pts⚠ The schedule says "Wed or Sat noon"
12/06SunPS4Final23:59 · 35 pts

Wrap-up · Wikipedia article and final report

12/7 – 12/21 · Wiki draft 5 · wiki peer review 9 · wiki final 25 · final report 30; Symposium 4

12/07Mon#26Work time for the project and Wikipedia articleProject and Wikipedia article work timeWork time
12/09Wed#27Work time for the project and Wikipedia articleProject and Wikipedia article work timeDue todayDWAWiki draft23:59 · 5 ptsWork time
12/11FriCR4PS4 meeting feedback checklist23:59 · 2 pts⚠ The schedule says 12/11, Canvas says 12/22; submit by the earlier one
12/14Mon#28Symposium 4 presentationsDue todaySP4Oral presentation12:30 · 5 ptsSR4Presentation peer review23:59 · 3 ptsPresentation
12/16Wed#29Where bioinformatics goes next; course evaluation (last class)Bioinformatics looking forward & course evaluation
12/17ThuWiki peer reviews ×3WR1–323:59 · 3 × 3 pts
12/18FriFWAWiki final, published17:00 · 25 pts
12/20SunFGRFinal gene report17:00 · 30 pts⚠ Canvas says 23:59, the schedule says 17:00; going with 17:00
12/21MonCS2End-of-course survey12:01 · extra credit

Gene and locusCoordinates from NCBI Gene and UCSC (hg38 RefSeq Select), all 1-based

Chromosome 1. LACTBL1 is at 1p36.12, near the end of the short arm, about 22.95 Mb.

The surrounding 680 kb. The nearest gene on the telomeric side is EPHB2 (EPH receptor B2), about 31.5 kb away. On the centromeric side come TEX46 (testis expressed 46), KDM1A (lysine demethylase 1A) and LUZP1 (leucine zipper protein 1), in that order. Hover over a gene to see its coordinates.

Exon–intron structure, to scale. Thin boxes are 5′UTR, thick boxes are coding. The start codon is in exon 2. Exon 8 is the longest (982 bp) and holds over half of the coding region.

Exon	Genomic coordinates (chr1)	Length	mRNA position	Following intron
E1	22,972,772–22,972,866	95 bp	1–95	340 bp
E2	22,972,305–22,972,431	127 bp	96–222	6,879 bp
E3	22,965,290–22,965,425	136 bp	223–358	2,073 bp
E4	22,963,107–22,963,216	110 bp	359–468	3,007 bp
E5	22,959,942–22,960,099	158 bp	469–626	1,021 bp
E6	22,958,685–22,958,920	236 bp	627–862	3,258 bp
E7	22,955,321–22,955,426	106 bp	863–968	1,296 bp
E8	22,953,043–22,954,024	982 bp	969–1950	—

TranscriptNM_001289974.2 · MANE Select · VALIDATED RefSeq

mRNA structure. Exons alternate dark and light; thin boxes are 5′UTR. The two bars below show where the domains fall on the mRNA.

The start

Start codon at nt 211, Kozak context agaATGa: A at −3, so strong; +4 is A, not G Record
The 5′UTR has one uORF (nt 197–214, MPWRE*); its stop codon overlaps the main ATG Predicted
No in-frame upstream stop codon in the 5′UTR, so it cannot be shown the 5′UTR is complete; the record only says "complete on the 5′ end" Record

The end

The record ends at the TGA stop codon (nt 1948–1950): no 3′UTR, no annotated polyA signal or site Record
In the genome there is an AATAAA about 323 nt downstream of the stop codon; only a candidate, needs PolyASite or EST evidence Predicted
The whole record is inferred from genomic alignment (derived from AL035704.9); there is no independent cDNA clone Record

mRNA sequence (FASTA)1,950 nt

>NM_001289974.2 Homo sapiens lactamase beta like 1 (LACTBL1), mRNA
ATTCCTGTCCTCTGGAACACAGGGGACTCCCAGAGGCACTGTCAGCCAAACTGGGGCCCT
AACTGCCCTGTTCTTGCTGGTACCTTCTGGAGAGGCCCCTCAGGACCTCTTTTTGCCTTC
ACTCCTCTGACTCAGTTTCCTGCTTACGGGATCTCCTGTGGTGTAGACCTGAATCCTCAA
AGAGCCAGAATCTTCCATGCCTTGGAGAGAATGAAGACCCAGGATGGCTGGTGGCCCAAC
TTGCTGAAAGTGAAGAAGAAGTGGCTCCTCCCGGCCATCTGCAGCTTCTTCTGCCTGCTC
TCAGTGGTCATGACTGGCTGCTTCCTGTGGCAGTATCACCTCCCCAAGCTGAAGACCGGT
TCCCTGGGACCAGAGGAGACCTCTGCCCCTGTGAGGATGTGTCCCCGGCACCCTGAGCCC
GTGCCCCTGGCTCACCCACTCCCCGTGTTGAAGGAGGCCCTGGAAAAGGTGGACCAGATC
CTGCGCCAGGCAATGTCTGCCCCAGGCGTGGCTGCCATGTCTGCAGTTGTCATCCACAAT
GATACTGTGCTCTGGACAGGGAACTTTGGGAAGAAGAATGGCTCAGACCCGGCTTCTGGG
GCCCCCAATGAGTACACCATGTACAGGATCTCCAGCATCTCCAAGATCTTTCCTGTCCTC
ATGCTGTACCGTCTGTGGGAGGAGGGCATCGTGGCCTCCCTAGATGACCCTCTGGAGCGG
TATGCCAGCACCTTCACCATTAACAACCCGCTGGGCCTGGCATCAGCCGAACAGCAGGGC
CTGATGGACGGGCTGGAGCAGGTGGGCCCCGCCCCAAGGCCTTCACCTGTCACCCTTCGA
AGGATGGCCAGCCAGCTCTCAGGGCTGCCCAGAAGGCTGCGCTCCACTTCACTGCTGTGG
AAGGGCAGCACCCAGGAGGCCCTGAACCTGCTCAAGGACGATGTGCTGGTAGTGGACCCG
GGAACCAGATGCCATTACAGCACGCTGGCCTTCTCGCTCCTGGCCCACGTCCTGGCAGCT
CACACCGCCCAGGGCGACTACCAGCGCTGGGTCTCGGAGAACGTGCTGGAGCCGCTGGGG
ATGGCAGACACGGGCTTTGACCTCACGCCCGACGTGCGCGCGCGCCTGGCCGCGGGCTTC
TACGGCAGCGGGCGGCCGGCGCCACTCTATGACCTGGGCTGGTACCGGCCGTCGGGCCAG
ATGTACTCTACCGCCGCCGACCTGGCCAAGCTGGCCGTGGCGCTCCTGGGCGGCGGGCCC
CGGCGGCTCCTGCGGCCCGACGCGGCCAAGACGCTGCTGGCGCCGCTGCTGGCCTGCCCG
GGCGCCTACTTCGCCAACGAGACGGGCACGCCGTGGGAGTTCCACGCGCAGCGGGGCTAC
CGCGTGGTGCGCAAGGACGGCGATCTGGACGGCTACGCCGCCACCTTCTCGCTGGTGCCC
CCGCTGCGTCTGGGCCTGGTGCTGCTGCTGGCAGGGCCACGGCCGCCCGGGCCCGACCTG
GTGGCGCGGGCCTACGATGAGCTCCTGCCCGCCCTGGAGCGCGCCCTCCGGGAGGCCGAG
CCCGGCCCGGCTCCGCCGCCCACCGCGCACCCCTTCGCCGGCTACTTCACCTTCGCCAAC
CTGACCTTCTACGAGGTGCGCGCCGGGCCGGCGGGCGAGCTGCGCCTGCGCCAGTTCGGG
CCGCGCGTGGAGGCGCTGGTGCCTCCAGCGTTCCGCACCCTGGCGCTGCGCCACCTGCAC
GGCCGCGTCTTCCAGCTGCACGTGGCCCACGAGTTCCCCTGCGCACTGCCACTCGGCGAC
GCGTGGCTCTCGCTCGAGGCCCAGCACGGGCAGCTGGTCAACTTCTACCCCTTGGATCAC
CACGGGCTGTCACCCGGCTTCGACGTGCCCGGCCTCAACACGTACAGAGTGCTGCGGCTG
CGGGGCAAGCCGGTGTTCAAGACCCAGTGA

ProteinNP_001276903.1 · UniProt A8MY62 (BLML_HUMAN) · 579 aa

Domain map and human–mouse identity. Solid bars are database annotations. Hatching and the diamond are inferred from sequence, with no experimental support. The curve below comes from the 9/21 Needle alignment, using a 15-residue sliding window. Divergence is highest around aa 186–200, where mouse has an insertion.

ProtParam values

Molecular weight	63,506.47 Da (~64 kDa)
Theoretical pI	9.00
Charged residues	Asp+Glu 50 · Arg+Lys 58
Most common amino acids	Leu 14.9% · Ala 10.5% · Pro 9.7% · Gly 8.3%
Extinction coefficient (280 nm)	89,185 M⁻¹cm⁻¹
Instability index	43.06 (classed as unstable)
GRAVY	−0.069

Source: ExPASy ProtParam, A8MY62. Does not account for post-translational modifications.

How far the evidence goes

UniProt protein existence level PE1 (detected by proteomics), but no function annotation and no signal peptide or transmembrane segment annotated Record
The AlphaFold2 model closely resembles a bacterial D-aspartyl endopeptidase (paenidase); the authors expressed the 63–450 fragment in E. coli, it was insoluble, and no activity was measured Literature
SISK at S142–K145 fits the serine beta-lactamase catalytic motif S-x-x-K and is conserved in mouse Predicted
Interactions: PRC1 (affinity purification MS) and ZC3HAV1 (proximity labeling MS), each from a single high-throughput dataset Literature
HPA subcellular localization to nuclear bodies rests on a single antibody (HPA066810) To check

Protein sequence (FASTA)579 aa

>NP_001276903.1 beta-lactamase-like protein 1 [Homo sapiens]
MKTQDGWWPNLLKVKKKWLLPAICSFFCLLSVVMTGCFLWQYHLPKLKTGSLGPEETSAP
VRMCPRHPEPVPLAHPLPVLKEALEKVDQILRQAMSAPGVAAMSAVVIHNDTVLWTGNFG
KKNGSDPASGAPNEYTMYRISSISKIFPVLMLYRLWEEGIVASLDDPLERYASTFTINNP
LGLASAEQQGLMDGLEQVGPAPRPSPVTLRRMASQLSGLPRRLRSTSLLWKGSTQEALNL
LKDDVLVVDPGTRCHYSTLAFSLLAHVLAAHTAQGDYQRWVSENVLEPLGMADTGFDLTP
DVRARLAAGFYGSGRPAPLYDLGWYRPSGQMYSTAADLAKLAVALLGGGPRRLLRPDAAK
TLLAPLLACPGAYFANETGTPWEFHAQRGYRVVRKDGDLDGYAATFSLVPPLRLGLVLLL
AGPRPPGPDLVARAYDELLPALERALREAEPGPAPPPTAHPFAGYFTFANLTFYEVRAGP
AGELRLRQFGPRVEALVPPAFRTLALRHLHGRVFQLHVAHEFPCALPLGDAWLSLEAQHG
QLVNFYPLDHHGLSPGFDVPGLNTYRVLRLRGKPVFKTQ

ExpressionThe two datasets use different units and cannot be compared directly

Human Protein Atlas

3.6nTPM · retina (tissue enriched)

33.9nCPM · retinal bipolar cells (single cell type enriched)

RNA data Record. Subcellular localization to nuclear bodies uses only one antibody, so take it with a grain of salt. Open HPA

GTEx (54 tissues)

0.73TPM · testis

0.06TPM · pancreas

Median is 0 in all other tissues Record. GTEx has no retina samples, so this does not contradict HPA. Open GTEx

Co-expression

Indrischek et al. 2022 (eLife) found that LACTBL1 expression clusters with phototransduction and vision-related genes Literature. This is an expression correlation, not functional evidence.

EvolutionPS2 analysis, 10/2–10/4; methods and all numbers are in PS2 draft v33

LACTBL1 is an old gene. Every major vertebrate group has it, and tunicates have a one-to-one ortholog, so the chordate ancestor 600 million years ago already carried it. Its split from the paralog LACTB is even older: sea anemones have both, so the duplication happened before cnidarians and bilaterians separated (about 680 million years ago). Among bilaterians it was found only in chordates; the nematode and fruit fly branch lacks it.

Group	Status	Representative species · identity to human (divergence time)
Mammals	Present	mouse 242707 81.1% (84 MYA) · opossum 72.8% · platypus 66.4% (181 MYA)
Reptiles and birds	Present	chicken 769085 62.3% · green anole 64.3% · painted turtle 64.2% · American alligator 60.8% (all 319 MYA)
Amphibians	Present	western clawed frog 116410950 59.1% · axolotl 61.0% (352 MYA)
Lobe-finned fish	Present	West African lungfish 122792879 55.8% · coelacanth 58.3% (402–413 MYA)
Ray-finned fish	Present	zebrafish lactbl1a 568717 50.6% · spotted gar 56.7% (435 MYA). None appear in the NCBI Datasets ortholog table; filled in from Ensembl and BLAST O2
Cartilaginous fish	Present	thorny skate LOC116990427 55.1% · elephant shark 56.8% (463 MYA)
Lampreys (jawless)	Present	river lamprey 144936302; the sea lamprey model is incomplete (flagged LOW QUALITY), and its 29.6% identity is pulled down mostly by gaps
Tunicates	Analysis	Ciona intestinalis XP_078489512.1, 29.8% (601 MYA). Passes reciprocal best hit, and Ensembl also calls it a one-to-one ortholog. The most distant strict ortholog
Sea anemone (cnidarians)	Analysis	XP_020906969.1, 30.2% (682 MYA). The sea anemone has 8 genes in the same subfamily and the second-best score is very close, so this counts as a one-to-many co-ortholog; there is also 1 LACTB-type gene
Sponge	Analysis	XP_003387189.1, 26.8% (750 MYA). Leans toward LACTBL1, but the sponge has no LACTB-type gene to compare against, so it cannot be decided
Protostomes, echinoderms, hemichordates, amphioxus	Not found	Not in the top 1,000 BLAST hits against invertebrate RefSeq; best tblastn hit in the amphioxus genome is E = 5.4. Whether it was lost or diverged beyond recognition was not tested
Protists, fungi, bacteria	Same family	Dictyostelium, fungi, and 757 bacterial proteins all hit (E ≤ 10⁻¹³). These are homologs from the whole beta-lactamase family, not orthologs

Identity: global alignment, EBLOSUM62, gap 10/0.5, gaps included in the denominator (first reproduced the six PS1 Needle runs, with identical results). Divergence time: TimeTree 5 median. NCBI Datasets lists orthologs in 553 species (10/2), 2 more than at the 9/13 download.

Paralog LACTB

E = 6×10⁻⁸phmmer local hits on two beta-lactamase domain segments

Full-length global identity is only 15.9%, indistinguishable from shuffled-sequence controls (11.5 ± 4.6%); only HMMER picked it up. BLAST against all human RefSeq hits only LACTBL1 itself. GLS and GLS2 share only the fold, with no sequence hits; LACTB2 has a similar name but is a metallo-beta-lactamase, not a homolog.

Rate of evolution

0.62×10⁻⁹substitutions / site / year (fit on 16 jawed vertebrates)

Cytochrome c is 0.20 and fibrinogen alpha 1.39; LACTBL1 sits in between. m = −ln(1 − p), straight-line fit through the origin. All three mammals sit high, and so do both reference proteins, so the problem is with the mammal points, not LACTBL1 itself.

LACTB duplication time

≥ ~900 Myrrough estimate 0.92–1.01 Gya, usable only as a lower bound

Human LACTBL1 vs LACTB: p = 0.684, m = 1.15, divided by each gene's own rate. At p this high, the Poisson correction underestimates m. Consistent with the sea anemone having both genes: older than 680 million years.

Most conserved 5%

29 residues7 are invariant in all three alignments

S142, K145, T208, G218, Y256, S333, D337 are fully invariant. 28 of the 29 fall in the Pfam domain (87–435) and line up with the catalytic center of E. coli AmpC (S142-x-x-K145 ↔ Ser80/Lys83). None are in the N-terminal transmembrane segment. So it has kept the fold, and possibly the active site; the sequence cannot tell us the substrate.

Two unrooted trees

The gene tree (17 strict orthologs, NJ, 1,000 bootstraps) recovers 6 groups from the species tree with 95–100% support. Disagreements are all on short branches with support below 85%, with no sign of paralogs in the ortholog set. The tunicate gets pulled next to the mammals, probably long-branch attraction. The time-calibrated tree uses TimeTree to date 16 nodes.

Three multiple sequence alignments

Done locally in FAMSA: 17 strict orthologs, 8 distant homologs (including bacterial AmpC), and 12 paralogs (LACTBL1 and LACTB from 6 species each). Domains, exon boundaries, and a consensus row are marked. To be redone after the 10/5 and 10/7 classes following the slides and Clustal Omega.

Ortholog table: human + 20 species (PS2 Table 4)Expand

Species	Group	MYA	Accession	aa	Identity	Similarity
Human	Primates	—	NP_001276903.1	579	100	100
Mouse	Rodents	83.5	NP_001230191.1	581	81.1	86.4
Gray short-tailed opossum	Marsupials	159.2	XP_007491271.2	584	72.8	80.7
Platypus	Monotremes	181.2	XP_028922027.1	588	66.4	75.8
Green anole	Squamates	318.8	XP_062821609.1	568	64.3	77.2
Western painted turtle	Turtles	318.8	XP_065431380.1	588	64.2	76.7
Chicken	Galliformes	318.8	XP_040507344.1	596	62.3	73.4
American alligator	Crocodilians	318.8	XP_059572258.1	616	60.8	72.1
Axolotl	Salamanders	352.1	XP_069466964.1	590	61.0	74.5
Western clawed frog	Frogs	352.1	XP_031758511.1	587	59.1	71.3
West African lungfish	Lungfish	402.2	XP_043916373.1	527	55.8	68.8
Coelacanth	Coelacanths	412.5	XP_014353639.1	567	58.3	72.0
Spotted gar	Holosteans	435.1	XP_069039981.1	601	56.7	71.1
Zebrafish	Teleosts	435.1	XP_087129935.1	653	50.6	64.1
Elephant shark	Holocephalans	463.2	XP_007905476.1	566	56.8	70.7
Thorny skate	Skates and rays	463.2	XP_032903974.1	555	55.1	69.3
Sea lamprey †	Jawless	562.8	XP_032824006.2	385	29.6	38.8
Ciona intestinalis	Tunicates	601.0	XP_078489512.1	525	29.8	46.8
Sea anemone Aiptasia	Cnidarians	682.3	XP_020906969.1	551	30.2	45.9
Sponge Amphimedon	Sponges	750.1	XP_003387189.1	535	26.8	41.7

Identity and similarity are %, from global alignment against human NP_001276903.1. † Partial model. Sequences pulled from NCBI Protein on 9/28.

Literature and associationsNo paper has LACTBL1 in its title yet; everything below mentions it in passing

10/5 follow-up: Google Scholar returns about 82 results for "LACTBL1", but I only read page 1 (paging was blocked by a CAPTCHA). USPTO Patent Public Search returns 67; 31 are one family of protein-degrader patents where LACTBL1 appears only in a long gene list. None is about LACTBL1.

Ito & Utsunomiya-Tate 2025, J Biol ChemPMID 40286848Literature AlphaFold2 structure comparison: LACTBL1 is structurally similar to a bacterial D-aspartyl endopeptidase (paenidase), with the key Tyr conserved. The recombinant fragment was insoluble and activity was not tested. The strongest structural lead so far.
Indrischek et al. 2022, eLifePMID 35727138Literature Expression pattern clusters with phototransduction and vision genes. The paper calls it a "putative serine proteinase" but cites a LACTB paper for it.
Capalbo et al. 2019, Nat CommunPMID 31586073Literature Interactome of cytokinesis intermediates; affinity purification mass spec pulled down PRC1. High-throughput, single data point.
Zheng et al. 2025, Sci DataPMID 40057523Literature TurboID proximity labeling of the PARP family gives a ZC3HAV1 interaction. The 3rd paper NCBI Gene currently links.
Gregory et al. 2006, NaturePMID 16710414Sequence and annotation of chromosome 1. LACTBL1 is included only because it sits on this chromosome.
GWAS Catalog: rs10158878GWASLiterature SNP about 20 kb upstream of the gene, associated with refractive error, p = 5 × 10⁻¹⁰. This is a locus association, not evidence for LACTBL1's own role.
Wang et al. 2022, World J Mens HealthPMID 35021295Literature Premature ejaculation GWAS in 486 Han Chinese men; 3 LACTBL1 SNPs reach only the suggestive threshold (p < 5 × 10⁻⁶), with no replication. All three are in intron 4, less than 1.2 kb apart, so they are likely the same signal.
Kitamura et al. 2023, Sci RepDOILiterature lactbl1 is higher in spontaneous tumors of the western clawed frog; in TCGA kidney cancer the high-expression group has worse survival. "Significant" there means a median split with a 0.2 difference, with no statistical test, and it conflicts with HPA's not prognostic. Weak evidence.
Zhao et al. 2025, Nat CommunPMID 40121201Not about LACTBL1 An Alzheimer's disease hydroxymethylation region that actually falls inside EPHB2; LACTBL1 is just a neighbor within 50 kb.
Liobikas et al. 2006, Protein Expr PurifPMID 16202624About LACTB, not LACTBL1. Often miscited, kept here for reference.

Still to checkAs of 10/5; the PS1 items are done or written into the final

To check NCBI Gene keyword search hit counts: LACTBL1 and "lactamase beta like 1" (the only blank left in PS2 1.1)
To check How many transcripts the sponge LOC100637838 NCBI Gene page lists (PS2 3.1; UniProt and KEGG both have only 1)
To check MSA, tree building, the formula for m and the fitting method, checked against the 10/7, 10/12, 10/14 slides
To check Expression question (if this year's PS2 has one): do after the 10/19 and 10/21 classes
To check Instructor feedback from the PS1 meeting (10/5–10/9); PS2 question 4 concept translations need revising based on it
To check Where the 3′ end is: the AATAAA at +322 nt past the stop codon sits right next to an Alu, with no transcript evidence; needs PolyASite or 3′ RACE
To check Whether the nematode and fruit fly branch lost LACTBL1 or it diverged beyond recognition
To check Whether the N-terminal hydrophobic segment is a signal anchor or a transmembrane segment: DeepTMHMM, SignalP (covered in PS3)
To check If I write about the 1p36 deletion CNV, find out how much DNA is deleted and how many genes go with it
Done Invertebrate orthologs (10/2) · Needle for chicken, zebrafish, tunicate (PS1 Q3) · Google Scholar and USPTO (10/5) · poly(A) answer written into PS1 v7

OverviewWhat can be confirmed from databases, literature and the PS2 analysis as of 2026-10-05

Data portal66 links, each pointing straight to a page for this gene, transcript or protein

NCBI core records

NCBI search and analysis

Genome and nomenclature

Protein and structure

Expression

Variants, disease and mouse

Literature and patents

Homologs (for Q3)

Course tools

Course scheduleAll 29 classes and every due date for the semester, in Chicago time. Dates checked against Canvas and the course schedule on 9/30. Open a past class to see my notes: what was covered and what I did with LACTBL1. Slides are on Canvas (UMN login required)

Round 1 · Gene identity, sequence, alignment

What was covered

What I did

What was covered

What I did

What was covered

What I did

What was covered

What I did

What was covered

What I did

What was covered

What I did

What was covered

What I did

Round 2 · Evolution and expression

Round 3 · Regulation, RNA, protein primary structure

Round 4 · Localization, structure, variants, interactions

Wrap-up · Wikipedia article and final report

Gene and locusCoordinates from NCBI Gene and UCSC (hg38 RefSeq Select), all 1-based

TranscriptNM_001289974.2 · MANE Select · VALIDATED RefSeq

The start

The end

ProteinNP_001276903.1 · UniProt A8MY62 (BLML_HUMAN) · 579 aa

ProtParam values

How far the evidence goes

ExpressionThe two datasets use different units and cannot be compared directly

Human Protein Atlas

GTEx (54 tissues)

Co-expression

EvolutionPS2 analysis, 10/2–10/4; methods and all numbers are in PS2 draft v33

Paralog LACTB

Rate of evolution

LACTB duplication time

Most conserved 5%

Two unrooted trees

Three multiple sequence alignments

Literature and associationsNo paper has LACTBL1 in its title yet; everything below mentions it in passing

Still to checkAs of 10/5; the PS1 items are done or written into the final